The AI Revolution

The AI Revolution

• 1. The AI Revolution Sri Viswanath Vibhor Khanna Yijia Liang Coatue November 2023

• 2. 2 Disclaimer THIS PRESENTATION DOES NOT CONSTITUTE AN OFFER TO SELL OR THE SOLICITATION OF ANY OFFER TO BUY AN INTEREST IN A FUND OR VEHICLE MANAGED BY COATUE MANAGEMENT, L.L.C. (“COATUE”). ANY SUCH OFFER OR SOLICITATION WILL BE MADE ONLY AT THE TIME A QUALIFIED OFFEREE RECEIVES A CONFIDENTIAL PRIVATE OFFERING MEMORANDUM OR OTHER OFFERING DOCUMENT (“CPOM”) DESCRIBING THE OFFERING AND RELATED SUBSCRIPTION AGREEMENT. IN THE CASE OF ANY INCONSISTENCY BETWEEN THE DESCRIPTIONS OR TERMS IN THIS PRESENTATION AND THE CPOM, THE CPOM SHALL CONTROL. COATUE HAS PROVIDED THIS PRESENTATION SOLELY IN CONNECTION WITH ONGOING OR INTENDED DISCUSSIONS WITH THE PERSON TO WHOM IT HAS BEEN DELIVERED. IN THIS REGARD, THE PERFORMANCE FIGURES, AND CERTAIN OTHER INFORMATION INCLUDED IN THIS PRESENTATION, REQUIRE FURTHER EXPLANATION AND MUST BE DISCUSSED WITH YOUR COATUE REPRESENTATIVES. ACCORDINGLY, PROSPECTIVE INVETORS SHOULD NOT RELY ON THE INFORMATION IN THIS PRESENTATION ABSENT SUCH DISCUSSIONS. NO SECURITIES OR SERVICES SHALL BE OFFERED OR SOLD IN ANY JURISDICTION IN WHICH SUCH OFFER, SOLICITATION OR SALE WOULD BE UNLAWFUL UNTIL THE REQUIREMENTS OF THE LAWS OF SUCH JURISDICTION HAVE BEEN SATISFIED. WHILE ALL THE INFORMATION PREPARED IN THIS PRESENTATION IS BELIEVED TO BE ACCURATE, COATUE MAKES NO EXPRESS WARRANTY AS TO THE COMPLETENESS OR ACCURACY NOR CAN IT ACCEPT RESPONSIBILITY FOR ERRORS, APPEARING IN THE PRESENTATION. NEITHER COATUE NOR ITS AFFILIATES ASSUMES ANY DUTY TO UPDATE THE INFORMATION CONTAINED HEREIN FOR SUBSEQUENT CHANGES OF ANY KIND AND THERE IS NO ASSURANCE THAT THE POLICIES, STRATEGIES OR APPROACHES DISCUSSED HEREIN WILL NOT CHANGE. ANY PROJECTIONS, MARKET OUTLOOKS OR ESTIMATES IN THIS PRESENTATION ARE FORWARD-LOOKING STATEMENTS AND ARE BASED UPON CERTAIN ASSUMPTIONS. OTHER EVENTS WHICH WERE NOT TAKEN INTO ACCOUNT MAY OCCUR AND MAY SIGNIFICANTLY AFFECT INVESTMENT RETURNS OR PERFORMANCE. ANY PROJECTIONS, OUTLOOKS OR ASSUMPTIONS SHOULD NOT BE CONSTRUED TO BE INDICATIVE OF THE ACTUAL EVENTS WHICH WILL OCCUR. HISTORICAL RETURNS ARE NOT PREDICTIVE OF FUTURE RESULTS. THE INFORMATION PROVIDED HEREIN, INCLUDING, WITHOUT LIMITATION, INVESTMENT STRATEGIES, INVESTMENT RESTRICTIONS AND PARAMETERS, PROCEDURES, POLICIES, AND INVESTMENT AND OTHER PERSONNEL MAY BE CHANGED, MODIFIED, TERMINATED OR SUPPLEMENTED AT ANY TIME, WITHOUT NOTICE. THIS PRESENTATION IS CONFIDENTIAL AND NOT INTENDED FOR PUBLIC USE OR DISTRIBUTION. THIS PRESENTATION HAS NOT BEEN APPROVED BY THE U.S. SECURITIES AND EXCHANGE COMMISSION OR ANY OTHER FEDERAL OR STATE REGULATOR. YOU SHOULD CAREFULLY READ THE IMPORTANT DISCLOSURES IN THE APPENDIX.

• 3. 3 Where we are in AI today Key Topics

• 4. We’re at Day 1 of AI…and riding on top of past waves 4 Source: World Bank as of September 2023, Statista as of September 2023, Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. % US Technology Adoption 1990 1995 2000 2005 2010 2015 2020 2025 25% 50% 75% 100% Internet Smartphones AI

• 5. Source: Bloomberg; Coatue opinion and analysis as of October 2023. AI Basket represents GS’s AI basket. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. In the first half of 2023, the AI ecosystem exploded! (1/2) ChatGPT weekly unique visits NVDA 2Q Earnings AI Basket Performance YTD 5 12/22 01/23 02/23 03/23 04/23 05/23 06/23 0M 10M 20M 30M 40M GPT-4 Released GPT-3.5 Released Only 5 months! $7B $11B Street estimates Actual guide +60% YTD 100 1/23 3/23 6/23 9/23

• 6. 2021 2022 2023 0 1,500 3,000 4,500 6,000 7,500 9,000 Source: GitHub as of June 2023, ThereIsAnAIForThat.com as of September 2023, Hugging Face as of September 2023, and Coatue analysis and opinion as of September 2023. Data visualized begins in 2021. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable statements or trends; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. In the first half of 2023, the AI ecosystem exploded! (2/2) % of trending Github repos in AI/ML New AI applications over time Number of models on Hugging Face 340K+ 8K+ 6 2021 2022 2023 0K 50K 100K 150K 200K 250K 300K 350K 2021 2022 2023 0% 10% 20% 30% 40% 50% 60% 50%

• 7. Adoption has been twice as fast with each platform shift 7 Source: Morgan Stanley Research as of September 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Halving of penetration time with new technology waves 20 Years 12 Years 6 Years 3 Years PCs Internet Mobile Gen AI ? # of years to reach 50% user penetration in the US

• 8. AI has potential to drive the economy for years to come 8 Source: Bloomberg, CapitalIQ, and Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See AppendixDisclosures for important disclosures, including regarding projections and forward-looking statements and trends. # of employees per $1M of revenue (inflation adjusted – S&P 500 companies) 7.8 6.4 5.1 5.3 5.1 <3? 1986 2000 2008 2019 Today Future w/ AI? ~35% Improvement in Productivity Driven by PC and Internet

• 9. We’ve seen massive investment in AI 9 Cumulative funding in private AI companies since 2020 Valuation before public product launch Source: Coatue analysis of Pitchbook data as of November 2023. Does not include funding raised before 1/1/2020. Companies included herein based on Pitchbook round data that Coatue categorizes as an “AI Company” which may include those with AI embedded in a product rather than it being their core competency, and includes all labeled companies in ecosystem, of which may include Coatue portfolio companies. This categorization is subjective and focused on “Gen AI” technologies, and may not be exhaustive of all companies that others may construe as Gen AI. Categorization could be impacted at any time by market factors, changes in laws and other factors. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable statements or trends; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. $0B $10B $20B $30B 01/20 01/21 01/22 01/23 01/24 $28.2B Open AI Raised $10B $1.0B $1.0B $4.0B $1.2B Series B Series A Series B Series B

• 10. 4 13 18 34 47 56 10 Coatue’s broader AI portfolio Source: Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Companies included herein based on current investments that Coatue categorizes as an “AI Company” which may include those with AI embedded in a product rather than it being their core competency. In addition, the original thesis for some of these companies may not have originally been about AI. Categorization could be impacted at any time by market factors, changes in laws and other factors. Positions in companies above may be held across multiple funds and are not indicative of the performance achieved in one specific fund. Coatue makes no guarantee regarding the completeness of this information; may not represent the entire universe of AI companies in the portfolio. See Coatue.com for a full list of Coatue portfolio companies. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. 2018 and earlier 2019 2020 2021 2022 2023 Stealth Co. Stealth Co. (Incubation) Stealth Co. Stealth Co. Stealth Co. # of Cumulative Investments in AI 50+ In AI Portfolio Including AI native companies as well as those with significant AI features and/or strategies

• 11. 11 We’ve backed AI founders across stages Source: Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Companies included herein based on current investments that Coatue categorizes as an “AI Company” which may include those with AI embedded in a product rather than it being their core competency. Categorization could be impacted at any time by market factors, changes in laws and other factors. Coatue makes no guarantee regarding the completeness of this information; may not represent the entire universe of AI companies in the portfolio. See Coatue.com for a full list of Coatue portfolio companies. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. 10 15 14 9 7 Pre-seed / Seed Series A Series B Series C Series D+ # of Coatue AI portfolio companies based on round when Coatue first invested Stealth Co. Stealth Co. Stealth Co. Stealth Co. Stealth Co.

• 12. 12 AI could break through the hype and improve our world Key Topics

• 13. 13 Coatue View: AI is not just hype Source: Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. • Comparison: Historic cycles like fiber and cloud represent two different outcomes for underlying infrastructure • Most investment in AI today is within the model layer, but it’s too early to declare who will be the AI model winners Value accrual misaligned with investment Value Accrual Characteristic of hype cycles Coatue view Historic Examples • Comparison: Self-driving cars finally arriving, but after 15+ years of work • AI is already useful within ~5 years. We are in early innings of adoption and expect models to improve, but AI regulation is likely and is a challenge to implement well Overestimating timeline & capabilities of technology Capabilities • Comparison: Quantum computing hype was promising in theory but has not yet proven widespread practical utility • AI already proving significant utility across domains Lack of widespread utility due to maturity of technology Value Accrual 1990s Fiber Autonomous vehicles Quantum computing

• 14. 14 Comparison: Some enabling technologies become a public good Source: 25iq as of September 2023, Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable statements or trends; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Fiber Infrastructure 1990s Telco Co’s raised $1.6T of equity & $600B of debt Bandwidth costs decreased 90% within 4 years Most of these companies no longer exist today… Value accrual Capabilities Utility

• 15. 15 …While cloud infrastructure became a huge market Source: Public filings, Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable statements or trends; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Cloud Infrastructure 2010s Entirely new computing paradigm Most were already public companies with resources to build out data centers Big 4 hyperscalers generate ~$150B+ in revenue annually Value accrual Capabilities Utility

• 16. 16 Where did value accrue in the cloud stack? Source: (1) Note the stack is not an exhaustive look at the cloud stack overall, and merely is a simplified selection of categories that broadly cover the cloud stack. Data on TAM is from Gartner as of April 2023 (including SaaS, PaaS, and IaaS categorizations and TAM) and Coatue analysis and opinion of Cloud Semis as of 2021. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable statements or trends; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Cloud Stack % of total in stack Example Companies Est. TAM 1 SaaS Apps PaaS IaaS Cloud Semis ~$260B ~$140B ~$200B ~$50B 40% 22% 30% 8% Value accrual Capabilities Utility

• 17. 17 In AI, funding concentrated in model layer for now Source: (1) Note the stack is not an exhaustive look at the AI stack overall, and merely is a simplified selection of categories that broadly cover the AI stack. Coatue analysis of Pitchbook data as of November 2023. Does not include funding raised before 1/1/2020. Companies included herein based on Pitchbook round data that Coatue categorizes as an “AI Company” which may include those with AI embedded in a product rather than it being their core competency. This categorization is subjective and focused on “Gen AI” technologies, and may not be exhaustive of all companies that others may construe as Gen AI. Categorization could be impacted at any time by market factors, changes in laws and other factors. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable statements or trends; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. AI Stack % of total in stack Example Companies 1 Total funding Apps ~$5B 17% Models ~$17B 60% AI Ops ~$1B 4% AI Cloud ~$4B 13% AI Semis ~$2B 6% Value accrual Capabilities Utility

• 18. 18 The jury is out on which model companies will win Source: Pitchbook data as of November 2023, Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable statements or trends; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. AI Models 2020s Private AI model companies raised ~$17B in venture funding since 2020 Open-source models are becoming more ubiquitous Outcome TBD ? Value accrual Capabilities Utility

• 19. 19 Comparison: AI advancing much faster than previous waves Source: 1) Mercedes received Level 3 ruling this year (alongside Cruise/Waymo L4 becoming mainstream this year, Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable statements or trends; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Levels of autonomy Autonomous vehicles Generative AI L5 Fully autonomous L4 Highly autonomous L3 Self-driving with light intervention1 L2 Tesla autopilot L1 Cruise control Superhuman reasoning & perception AI autopilots for complex tasks AI co-pilot for skilled labor Supporting humans with basic tasks Generating basic content 15 Years ~5 Years Value accrual Capabilities Utility

• 20. 01/2022 03/22 05/22 07/22 09/22 11/22 01/23 03/23 05/23 07/23 09/23 50M 100M 150M 200M 250M 300M 350M 400M 450M 500M 550M 11/23 Approximate global AI applications users count by month Has initial AI enthusiasm slowed down? 20 Source: Coatue analysis and opinion as of November 2023. Non-exhaustive set of apps represented in sample set.For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. AI usage worldwide started flattening during summer User growth has slowed since May Usage slowing could be many reasons • Summer break for students • Shock and awe waning, people know which use cases AI is great vs. poor at Vast majority of usage (over 60% of userbase) driven by ChatGPT Value accrual Capabilities Utility

• 21. 2015 2016 2017 2018 2019 2020 2021 2022 2023 5% 10% 15% 20% 25% 30% 35% Early developer excitement waning and washing out AI “tourists” 21 Source: GitHub as of November 2023, Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See AppendixDisclosures for important disclosures, including regarding projections and forward-looking statements and trends. Excitement in AI/ML on GitHub, measured by stars, has declined since April 2023 Value accrual Capabilities Utility Percentage of all new GitHub stars given to AI/ML repos Down 13 percentage points from peak!

• 22. Serious AI builders remain 22 Source: GitHub as of November 2023, Commit means a discrete change to one or more files in a GitHub repository. Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. AI/ML commits have not declined as much since April 2023 Value accrual Capabilities Utility 2015 2016 2017 2018 2019 2020 2021 2022 2023 2% 4% 6% 8% 10% Percent of AI/ML repository commits as proportion of total commits on GitHub AI/ML commits

• 23. ChatGPT usage has rebounded, new capabilities have shipped 23 Source: Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. ChatGPT weekly web visits worldwide since launch Value accrual Capabilities Utility 100M 200M 300M 400M 01/23 10/23 50M 150M 250M 350M 450M 500M 12/2022 02/23 03/23 04/23 05/23 06/23 07/23 08/23 09/23 11/23 +27% from late August! Could new capabilities such as GPT4Vision, Web Search or the GPTs marketplace continue driving ChatGPT usage forward?

• 24. We believe it’s first innings of enterprise AI adoption 24 ~60%+ of surveyed enterprises plan to adopt AI Less than 10% have fully adopted; timeline is long Source: Survey of ~600 enterprise executives as of 2023, and Coatue analysis and opinion as of November 2023 For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. 35% 39% 42% 65% 61% 58% AI-powered features for existing products New AI products AI personalization & support Adopting Not adopting Value accrual Capabilities Utility 9% 6% 9% 33% 20% 26% 22% 25% 21% 35% 49% 43% AI-powered features for existing products New AI products AI personalization & support Early testing / MVP In prod. / scale rollout Fully adopted POC or earlier

• 25. We are optimistic: AI is getting better, faster 25 Source: Directional data taken from Contextual.ai, Time, and PapersWithCode as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Speed for models to reach human level accuracy on benchmarks has decreased 1998 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 2022 2024 Human parity Handriting recognition Speech recognition Image recognition Reading comprehension Language understanding Common sense completion High school & college subjects Code generation Value accrual Capabilities Utility

• 26. AI regulation may be more likely than most think 26 Source: AI Policy Institute and YouGov as of November 2023, Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Initial polls suggest many Americans concerned about AI 83% 82% 57% 18% I believe AI could cause a catastrophic event AI should proceed slowly and deliberately Support a 6-month pause in AI development I trust AI tech executives to self-regulate AI % of surveyed respondents, AI Policy Institute Value accrual Capabilities Utility

• 27. GPT-4 Cohere Command Claude v1 LLAMA-1 PaLM-2 BLOOM Draft EU AI Act requirements Non-exhaustive list Data sources “Description of the data sources used in the development of the foundation model.” Compute “Description of the training resources used by the foundation model including compute required, training time, and other information related to the size and power of the model.” Energy “Design and develop the foundation model, making use of applicable standards to reduce energy use, resource use and waste, as well as to increase energy efficiency.” Capabilities & limitations “Description of the capabilities and limitations of the foundation model.” Risk & mitigations “The reasonably foreseeable risks and the measures that have been taken to mitigate them as well as remaining non-mitigated risks with an explanation on the reason why they cannot be mitigated.” Data copyright “Without prejudice to national or Union legislation on copyright, document and make publicly available a sufficiently detailed summary of the use of training data protected under copyright law.” EU’s AI Act is among first examples of regulation 27 Source: Stanford as of November 2023, Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See AppendixDisclosures for important disclosures, including regarding projections and forward-looking statements and trends. Most models fall short of meeting requirements from EU AI Act, per Stanford study Value accrual Capabilities Utility

• 28. Comparison: Is AI hype justified with real utility? 28 Source: Google Trends data of September 2023, news clippings as of September 2023, Coatue opinion and analysis as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. No indication or guarantee that quantum computing interest will continue to grow. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Quantum computing interest has steadily grown without real world proof points…what about Generative AI? 0 10 20 30 40 50 60 70 80 90 100 Google trend ranking for “Quantum computing” vs” Generative AI” 2004 2007 2010 2013 2016 2019 2022 May 2003 Value accrual Capabilities Utility Sep 2011 Jan 2019 IBM says first quantum computing system ‼ is available, but is not widely used Quantum computing Generative AI

• 29. AI is delivering game-changing value Time saved for developers using Github Copilot 55% 90% Time saved from editing video on Runway Reduction in humananswered customer support requests 45% 79% AI chat rated higher quality vs. physician responses Source: Github as of September 2022, Runway as of September 2203, Arxiv as of April 2023, JAMA as of April 2023, and Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable statements or trends; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. 29 Value accrual Capabilities Utility

• 30. AI poised to improve productivity significantly! Example: Designing Marketing Materials Design on paper Iterate on physical copies Design on local software Easy to iterate Design on web apps Instant collaboration AI-created design Human prompts with text Manual Physical Single-player Manual Digital Single-player Manual Digital Collaborative Automated Digital Collaborative Analog 1950s PCs 1960s – 1980s Internet & Cloud 1990s – 2010s Generative AI 2020s + Beyond Months Weeks Days Minutes Source: Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. 30 Value accrual Capabilities Utility

• 31. GitHub Copilot makes programming faster 31 Over 8M downloads of Copilot extension Copilot improving developer productivity over time Source: GitHub acceptance rates as of June 2023, and Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. 27% 28% 29% 30% 31% 32% 33% 34% Acceptance rate of Copilot suggestions over time Month 1 M2 M3 M4 M5 M6 Value accrual Capabilities Utility 1M 2M 3M 4M 5M 6M 7M 8M 9M 10M 11M 01/2022 07/22 01/23 07/23 01/24 Cumulative downloads of GitHub Copilot 10M+

• 32. 32 Source: Based on one company’s internal assessment data – not representative of all companies or guarantee other companies would achieve similar results. Customer satisfaction calculated by number of employees rating the interaction a 4 or 5 out of 5. Data compares against top 8 human employees, based on number of tickets participated in. Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. AI is cheaper customer support AI provides faster responses AI makes customers happier Value accrual Capabilities Utility $2.58 $0.13 Human AI -95% Cost per support interaction 1 Human AI 45+ 55% 69% Human AI Median response time (min) Median customer satisfaction Companies have seen huge efficiency gains already Experience from one Fintech company

• 33. Knowledge work, such as consulting, could be transformed by AI 33 Source: HBS & BCG study as of November 2023, Coatue analysis and opinion as of November 2023. Represents experience from one consulting firm in on particular study. Results may not be comparable across all consulting firms, and we are not making a statement that results would translate into different contexts and scenarios. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forwardlooking statements and trends. In one study, BCG consultants using AI performed better across all task metrics, including 40% better quality work Performance improvement on various tasks (1 being control group performance) 0.000 1.190 0.935 1.020 1.105 1.275 1.360 1.445 1.530 Task completion rate Task completion speed Task completion quality 1 1.12X 1 1.25X 1 1.40X Control With GPT-4 Value accrual Capabilities Utility

• 34. 34 We believe open-source is the lifeblood of AI Key Topics

• 35. 35 How we got here – AI is built in the open Source: Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable statements or trends; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. • Research from academia & industry has driven advancements in AI • Collaboration using open-sourced state-of-the-art models has led to rapid pace of innovation Openly Available Research Research • There are varying degrees of openness across AI today • Companies realize the value of their own data, and model providers have become more secretive Models & Data Models & Data • GitHub and Hugging Face underpin the open-source AI community • Participation in AI has been explosive Open Community Community

• 36. AI is a result of open research 36 Source: Arxiv and Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable statements or trends; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Cumulative AI/ML publications submitted on Arxiv 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 50K 100K Research Community Models & Data

• 37. Open collaboration accelerates innovation in AI 37 Source: Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable statements or trends; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Illustrative launches of AI models over time (non-exhaustive) GPT-1 announced on the heel of transformers Stable Diffusion launches LLaMA in Feb ‘23 launched on Aug ‘22 Blenderbot GPT-2 GPT-3 GPT4all Dreambooth GGML Falcon LLM LLaMA 2 launches in July ‘23 Vicuna WizardLM Research Community Models & Data 2018 2022 2023

• 38. The AI developer community has exploded! 38 Software developers are becoming AI engineers Hobbyists are getting involved Source: Developers measured by number of unique people making commits to AI repositories. Engagement measured by unique users adding stars, filing issues, etc. These may not match what GitHub tracks internally and should not be construed as a source of truth. Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Research Community Models & Data 2015 2017 2019 2021 2023 0K 25K 50K 75K 100K 125K 150K 175K 200K 225K # AI/ML Developers on GitHub # Users Engaging with AI/ML Projects on GitHub 2015 2017 2019 2021 2023 0M 1M 2M 3M 4M 5M

• 39. 39 GitHub has been a place to discover new AI/ML projects Cumulative GitHub stars given to AI/ML related projects 0M 5M 10M 15M 20M 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 Source: GitHub as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Research Community Models & Data

• 40. Open-source AI is inflecting on Hugging Face as well 40 Source: Tracks the quarterly averages of the number of downloads within last month for the Top 30 models on Hugging Face (ranked by number of downloads over the trailing 30 days). Coatue analysis and opinion of Hugging Face data as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Number of times AI models have been downloaded from Hugging Face Research Community Models & Data 50M 100M 150M 200M 250M 300M 350M 400M 2020 2021 2022 2023

• 41. However, AI is not always truly “open” (1/2) 41 Source: OpenAI GPT-3 and GPT4 papers as of September 2023, LLAMA-1 and LLAMA-2 papers as of September 2023, Google PALM details as of September 2023, Mosaic MPT model announcement as of September 2023, Information about each model taken from model publications and research papers. Models were selected based on broad popularity, across open source and proprietary large language models. Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable statements or trends; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Coatue’s open-source AI model checklist GPT-2 Feb 2019GPT-3 Jul 2020GPT-4 Mar 2023LlaMA Feb 2023 LlaMA-2 Jul 2023Mistral-7B Oct 2023 Dimensions of openness Model code Model weights Training data Model evaluation Architectural decisions Open commercial license Research Community Models & Data Meta did not release training data specifics for LlaMA-2. Could training data be the next battleground?

• 42. However, AI is not always truly “open” (2/2) 42 Source: OpenaI Codex paper as of September 2023, Salesforce Codegen paperas of September 2023, Replit Code model announcement as of September 2023, Dall-E 2 paper as of September 2023, Midjourney website as of September 2023, Stable Diffusion announcement as of September 2023. Models were selected based on broad popularity, across open source and proprietary code and image models. Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable statements or trends; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Coatue’s open-source AI model checklist Codex Aug 2021 Codegen Mar 2022 Code v1 May 2023 Dall-E 2 Apr 2022 Midjourney Jul 2022Stable Diffusion 1.0 Aug 2022 Dimensions of openness Model code Model weights Training data Model evaluation Architectural decisions Open commercial license Research Community Models & Data

• 43. Data is finally a new currency 43 Source: TheVerge as of January 2023, NewsAndMediaAlliance as of September 2023, 13NewsNow as of June 2023, Reuters as of July 2023, X as of September 2023, Coatue analysis and opinion as of November 2023. For illustrative purposes only. Logos listed above as examples of the applicable statements or trends; do not necessarily represent Coatue investments. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Timeline of companies cracking down on data access in 2023 January 2023 April 2023 July 2023 July 2023 • Announces lawsuit against Stability AI for infringing on copyright and trademark protections • Introduces new rules including mandatory negotiations with publishers for right to use content • CEO announces plans to begin charging AI developers for access to data on the platform • Reddit turns on API pricing in July. A popular app Apollo would need to pay an estimated $20M / year • X announces rate limits and cracks down on data scrapers $12K for 50M API calls $5K for reading 1M posts (monthly) Research Community Models & Data

• 44. Despite signs of “closed AI”, open-source models are catching up 44 Source: LMSYS as of November 2023, Coatue analysis and opinion as of November 2023. Non-exhaustive list of models from LMSYS, selected to show representative list across different model providers and sizes, and is not in order of LMSYS rankings. For illustrative purposes only. Logos listed above as examples of the applicable statements or trends; do not necessarily represent Coatue investments. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Chatbot rankings based on human feedback ELO Ratings of LLMs (50K+ votes) Closed-Source Open-Source LMSYS 1181 1155 1115 1092 1051 1046 1031 973 914 GPT-4 Claude-1 GPT-3.5 Turbo Vicuna-33B LLAMA-2 70B chat MPT-30B-chat Mistral-7B Instruct Koala-13B Alpaca-13B

• 45. 45 AI is transforming the tech ecosystem Key Topics

• 46. 46 The new AI-centric technology ecosystem Source: Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. AI models AI apps End users AI developer tools Cloud platforms Data center / hardware / power

• 47. 47 Foundation models are at the center of AI Source: Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable statements or trends; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. AI models AI apps End users AI developer tools Cloud platforms Data center / hardware / power AI models Models Cloud, DC, Semis Developer Tools Applications

• 48. 48 Last 50 years was about building faster & faster “calculators”… Computer or “Calculator” Serial processing “1 instruction at a time” Chips got smaller & more powerful 1500nm 100nm 28nm 3nm Based on instruction by programmer Follows sequential programming logic Does not require data CPU Software Source: Coatue opinion and analysis as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Models Cloud, DC, Semis Developer Tools Applications

• 49. 49 But next 50 years will be about building super-intelligent “brains” Brains AI Models Neural networks trained by data Learns patterns from data System makes decisions based on model rather than explicit instructions “Reasoning” is opaque, not driven by programming logic GPU Parallel processing Many calculations simultaneously Much more than calculators! Could become connected: Brain-to-brain network = AI Internet? Source: Coatue opinion and analysis as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Models Cloud, DC, Semis Developer Tools Applications

• 50. 50 AI enables a new platform: Intelligence-as-a-service Intelligence is the next layer of innovation Source: Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable statements or trends; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. “The right way to think of the models we create is a reasoning engine, not a fact database.” Future “[One of my] opinions on current LLMs: They are "reactive" & don't plan nor reason.” Today aa Softwareas-a-servicePlatformas-a-serviceInfrastructureas-a-service 1 Augment our capabilities 2 Automate our workflows 3 Change our existing behaviors IQaaS Intelligenceas-a-service Cursor Models Cloud, DC, Semis Developer Tools Applications GPTs

• 51. More research & headcount is not enough to win in AI 51 MSFT leads big research houses in publications OpenAI has shipped faster than larger peers Source: CSET as of September 2023, OpenAI as of September 2023, Coatue analysis and opinion as of November 2023. Deepmind headcount is of all employees based on Coatue tracking. Facebook AI headcount represents all Coatue labeled “ML” engineers at the company, not solely FAIR. OpenAI headcount represents all engineering headcount. OpenAI has shipped Chat-GPT, GPT-4, GPT-4Vision, Dalle-3 in last 12 months, compared to Google shipping Bard, Meta shipping LLAMA-1 and LLAMA-2 as far as major releases. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Models Cloud, DC, Semis Developer Tools Applications Microsoft Google Facebook AI OpenAI 10K 8K 3K 200 ~2,000 ~1,900 ~500 Google Deepmind Facebook AI OpenAI Number of AI/ML related research publications Number of AI/ML related full-time employees

• 52. Talent has historically been concentrated at a few AI model hubs 52 Source: Coatue analysis and opinion as of November 2023. For simplification, DeepMind includes Google Brain. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable statements or trends; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Major AI model providers are “poaching” talent from one another Models Cloud, DC, Semis Developer Tools Applications Talent inflows & outflows at major research hubs Illustrative only: Represents both direct & indirect (>1 hop) flows of talent INFLOW: 25 people joined OpenAI from Google DeepMind OUTFLOW: 18 people joined Google DeepMind from FAIR

• 53. Many talented people are now leaving to start new companies 53 Source: Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable statements or trends; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Example AI mafias Models Cloud, DC, Semis Developer Tools Applications Founders from: Number of founders from AI research houses Example companies founded: 36 32 27 15

• 54. 54 Scaling AI model performance has been a focus of AI researchers More parameters Will scaling parameters continue improving performance? GPT-4 still “king”; scaling experiments likely continuing Training & inference is getting optimized; AI at the edge is emerging Larger models need more data; extending data runway & optimizing quality crucial Larger datasets How much data is optimal for training models? More compute Can we reduce the compute costs of training & inference? Longer training Can training for more epochs improve performance? Still an open question Covered in next section on Cloud, DC, Semis Not covered; waiting for more data points to emerge Source: Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Models Cloud, DC, Semis Developer Tools Applications Strategies to scale Open research questions Coatue View ? Focus of remainder of section

• 55. Through 2022: We saw LLM parameters balloon! 55 Source: Arxiv as of September 2023, Hugging Face as of September 2023, Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Model names listed above are examples of the applicable AI models; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Illustrative view of parameter sizes of large language models through 2022 Models Cloud, DC, Semis Developer Tools Applications 01/2018 07/18 01/19 07/19 01/20 07/20 01/21 07/21 01/22 07/22 0.01 0.10 1.00 10.00 100.00 1,000.00 10,000.00 Parti Parameter size (B, log) LaMDA AlexaTM 20B Minerva (540B) GPT-NeoX-20B HuBERT Codex FLAN GPT-3 175B (davinci) ALBERT-xxlarge PaLM (540B) GPT-2 XLNet RoBERTa Large BART-large T5-11B GPT ELMo BLOOM Chinchilla BERT-Large = Relative model parameter size Moore’s Law: doubling every 2 years

• 56. 2023: GPT-4 remains “king”…can we continue scaling? 56 Source: LMSYS, Hugging Face, Semianalysis, Coatue analysis and opinion as of November 2023. Selected benchmarks are examples and are not exhaustive nor are they representative of objective “truth” about which models are better. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Model names listed above are examples of the applicable AI models; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Comparison of 2023 models across ELO ranking and MMLU benchmarks…GPT-4 is still the best Models Cloud, DC, Semis Developer Tools Applications = Relative model parameter size Higher ELO is better 900 950 1,000 1,050 1,100 1,150 1,200 1,250 0 10 20 30 40 50 60 70 80 90 100 ELO ranking (human feedback) Model performance on MMLU benchmark GPT-4 GPT-3.5 LLAMA-2 70B Vicuna-33B MPT-30B Vicuna-13B Vicuna-7B MPT-7B Mistral-7B Instruct

• 57. Coatue opinion and analysis as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. What if there was a $1B training-cost model? Model Training Cost Over Time 57 Model Training Cost Time Next-Gen Model Today GPT 4 GPT 3 GPT 2 $50K $100M $1B? $10M What would such a model be capable of? Models Cloud, DC, Semis Developer Tools Applications

• 58. How AI models are made: Data is a crucial component 58 Source: Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable statements or trends; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Data is upstream in process of developing good models Models Cloud, DC, Semis Developer Tools Applications Data prep Data evaluation & curation Model training Model deployment Public web data Custom domain data Proprietary data Training & deployment frameworks Compute providers Model evaluation Hosted inference Training infra (pre-training & fine-tuning) Select data to train model on, across various sources Clean & curate to improve dataset quality Create custom model architecture using training frameworks Train & iterate models to improve performance Deploy models & continuously evaluate to finetune models

• 59. Scaling models require scaling datasets 59 Source: Arxiv, Hugging Face, Meta, Semianalysis, PapersWithCode, CNBC, Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Model names listed above are examples of the applicable AI models; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Number of tokens in training datasets increasing overall 04/20 07/20 10/20 01/21 04/21 07/21 10/21 01/22 04/22 07/22 10/22 01/23 04/23 07/23 100 1,000 10,000 100,000 Training token dataset size (log, B) GPT-4 GPT-3 LLAMA-1 LLAMA-2 PALM PALM-2 BLOOM MPT-30B Code LLAMA GPT-4 rumored to have trained on 13T tokens! Models Cloud, DC, Semis Developer Tools Applications = Relative model parameter size

• 60. Data scarcity is a potential wall to scaling models 60 Source: Arxiv (Villalobos 2022), Coatue analysis and opinion as of November 2023. For illustrative purposes only; there is no guarantee that data will be exhausted or that image and video data will have a longer runway. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. High-quality text data could be exhausted soon, images & video have longer runway Illustrative: Amount of data collected relative to overall supply Models Cloud, DC, Semis Developer Tools Applications 2018 2020 2022 2024 2026 2028 2030 2032 2034 2036 2038 2040 2042 2044 2046 2048 2050 Probable that high quality text exhausted by 2026? Image + video data estimated to last until ~2040s, yet will be used to train different types of models High quality text Image + video Overall data supply (relative for each data type)

• 61. Data quality is just as important as data quantity 61 Source: Cohere as of September 2023, note that tested model sizes range from 124M to 1.5B. Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable statements or trends; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Emerging evidence that training on pruned datasets can result in similar performance in language models Performance on variety of language model benchmarks 78% 64% 65% 50% 79% 66% 67% 52% 0% 30% 60% 90% 10% 20% 40% 50% 70% 80% 100% Sentiment analysis Semantic similarity Text classification Natural language inference Full dataset Pruned dataset (30% of size) For up to a 1.5B parameter model, pre-training on just 30% of the dataset size can result in similar, if not better performance, across multiple tests! Can this generalize to even larger models? Models Cloud, DC, Semis Developer Tools Applications

• 62. Synthetic data can augment fine-tuning 62 Source: Arxiv (Tang 2023), Coatue analysis and opinion, as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Synthetically generated data can help clean or distill datasets for fine-tuning Models Cloud, DC, Semis Developer Tools Applications Results: Synthetic data fine-tuning yields significant improvement compared to zero-shot attempts, though is not yet comparable to real data Percentile performance on test 19 33 74 78 55 68 77 84 80 85 81 87 Named entity recognition (Precision) Named entity recognition (Recall) Relation extraction (Precision) Relation extraction (Recall) +196% +103% Zero-shot ChatGPT Fine-tuned on synthetic data (BERT) Fine-tuned on real data (BERT) Zero-shot ChatGPT already decent at some bio tasks!

• 63. 63 User feedback data is another way to improve performance Midjourney collects user feedback to continue refining models Source: AITuts, DIscord, Coatue analysis and opinion, as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Users choose which image to upscale (or generate a variant of), effectively choosing their preferred image Prompt /dungeons and dragons, female knight, of the rolling plains, full body, dark azure, victorian genre paintings, serene face, realistic depiction of light, golden light V5 (Mar 2023) Midjourney V1 (Feb 2022) V2 (Apr 2022) V3 (Jul 2022) V4 (Nov 2022) Models Cloud, DC, Semis Developer Tools Applications V5.2 (Jun 2023) Note: Midjourney improvement is partially due to collecting user feedback but is only one part of the equation!

• 64. There are many other opportunities to improve models! 64 Source: Coatue analysis and opinion, as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Beyond scaling parameters & data, there are many angles to push on Time Non-exhaustive, illustrative opportunity set Longer contexts Multi-modality Local models Better model eval New architectures Models Cloud, DC, Semis Developer Tools Applications

• 65. AI models AI apps End users AI developer tools Cloud platforms Data center / hardware / power AI has potential to re-accelerate underlying infrastructure Source: Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable AI company type; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends.

• 66. This AI wave has been extremely compute hungry 66 Source: Arxiv, Coatue analysis and opinion, as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. ML models are doubling compute needs for training in months Months to double FLOPS used in model training 21.3 Months 5.7 Months Pre-Deep Learning Era Deep Learning Era -73% 1952 - 2010 2010+ Training compute needs now grows faster than Moore’s Law Models Cloud, DC, Semis Developer Tools Applications

• 67. Training compute has exponentially increased with model sizes 67 Source: Epoch AI as of September 2023, Coatue opinion and analysis as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Model names listed above as examples of the applicable AI models; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Massive inflection in compute needs for AI, ~70X increase from GPT-3 to GPT-4 01/18 07/18 01/19 07/19 01/20 07/20 01/21 07/21 01/22 07/22 01/23 07/23 0.01 0.10 1.00 10.00 100.00 1,000.00 10,000.00 100,000.00 1,000,000.00 Computational Petaflops (M, log scale) Chinchilla - 70B GPT - 120M GPT-2 - 1.5B GPT-3 - 175B GPT-4 - 1.8T GPT-NeoX - 20B HuBERT - 375M Gato - 1.2B LLaMA - 65B ALBERT-xxlarge - 223M LaMDA - 137B Minerva - 540B AlphaCode - 41B NLLB - 55B AlexaTM - 20B BERT-Large - 345M Sparse all-MLP - 10B T5 - 11B T5 - 3B PaLM - 540B = Relative model parameter size Models Cloud, DC, Semis Developer Tools Applications Latest executive order requires robust reporting for models trained using more than 100B PFLOPs (10^26 FLOPs). This is ~5X larger than GPT-4’s rumored training compute of 21B PFLOPs

• 68. Inference compute likely to dramatically outpace training 68 Source: Semianalysis, Coatue opinion and analysis as of November 2023. Assumes 560 TFLOPS/token, 150 tokens per query. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Ongoing inference likely to require much more compute than one-time training 0 20 40 60 80 100 Computational Petaflops (B) GPT-4 Training 1M GPT-4 users 10M GPT-4 users 100M GPT-4 users 21B 1B 8B 80B +281% Ongoing daily compute cost! 1 Assuming 100M users at scale making 10 queries per day, that’s a ~4X increase in computational need incurred PER DAY One time compute cost! Models Cloud, DC, Semis Developer Tools Applications

• 69. AI compute has been primarily served by GPUs 69 Source: Lambda Labs, Inference Benchmark for Stable Diffusion as of 2023. Assuming single precision. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. GPUs are able to serve AI workloads much faster than CPUs Time to produce an image with Stable Diffusion (lower is better) 6 Seconds 8 Seconds 8 Seconds 9 Seconds 12 Seconds A100 RTX 3090 A6000 A5500 RTX 8000 i7-6850K EPYC 7352 459 Seconds 530 Seconds GPUs 70x faster! GPUs CPUs Models Cloud, DC, Semis Developer Tools Applications

• 70. The demand for GPUs has only begun 70 Source: Trendforce as of May 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Estimated ramp up for AI servers growing 25%+ Y/Y through 2026 Estimated global AI server shipments 855 1,183 1,504 1,895 2,369 2022 2023E 2024E 2025E 2026E +38% +27% +26% +25% What are the implications for this global shift in demand for GPUs? Models Cloud, DC, Semis Developer Tools Applications

• 71. 71 Follow the GPUs: AI poised to have impacts across our economy Source: NVIDIA public filings as of September 2023, Coatue analysis and opinion, as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclaimers for important disclosures, including regarding projections and forward-looking statements and companies and trends. Models Cloud, DC, Semis Developer Tools Applications Cloud Data Centers Power Semis $32B+ NVDA Revenue last 12 months

• 72. 72 The AI wave may stress our power grid Source: NVIDIA public filings as of September 2023, PTR Inc, University of Wisconsin, IEA, Schroders, Coatue analysis and opinion, as of November 2023. For illustrative purposes only. Potential power and price increases are Coatue estimates based on external data sources and are subject to change based on several factors, including demand, supply constraints, competition, the pace of AI adoption. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclaimers for important disclosures, including regarding projections and forward-looking statements and companies and trends. $32B+ NVDA Revenue last 12 months More data center power required ’23-24E & 3-5 GW 50% 2-3x 3-5 Years Electrical transformer price increases Potential increase in power demand by 2026 Leadtime to connect to the grid = Models Cloud, DC, Semis Developer Tools Applications

• 73. 73 More GPUs likely means more expensive servers Source: NVIDIA public filings as of September 2023, Semianalysis as of September 2023 showing server bill of materials. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclaimers for important disclosures, including regarding projections and forward-looking statements and companies and trends. $32B+ NVDA Revenue last 12 months Models Cloud, DC, Semis Developer Tools Applications Higher Server Bill of Materials Due to GPUs ’23-24E Note: all estimated numbers Non-AI Server AI Server CPU ~$2K 10x ~$20K GPU ~$200K Storage & RAM ~$5k 5x ~$20K Power Supply <$1K 10x ~$5K Networking & Other ~$5K 3x $15k Server Cost ~$10K 26x ~$260K

• 74. 2Q 3Q 4Q 1Q 2Q 3Q 4Q 1Q 2Q Could AI workloads drive a cloud reacceleration? Y/Y revenue growth rates for AWS, GCP, and Azure from Q2 2022 to Q2 2023 2022 2023E 2024E 36% 19% Source: NVIDIA public filings as of 2023, AWS & GCP public filings, Azure sell-side consensus from Bloomberg as of September 2023. Coatue analysis and opinion, as of November 2023. For illustrative purposes only. Illustrative future revenue growth rates for cloud providers are solely visuals and should not be interpreted as real figures. See Appendix-Disclaimers for important disclosures, including regarding projections and forward-looking statements. $32B+ NVDA Revenue last 12 months Cloud growth rates have decelerated over last four quarters… Models Cloud, DC, Semis Developer Tools Applications AI could be a catalyst for growth!

• 75. Pre 1990 1990 - 2006 2007 - 2014 2014 - 2022 2022 onwards ~$50B ~$100B ~$150B ~$250B ~$500B+ ? ~$50B ~$200B ~$350B ~$600B ~$1.5tn+ ? 75 Mainframe PC + Internet Smartphone Public Cloud AI Time Period Total Semi Spend Key HW players Source: SIA Industry data as of September 2023, Coatue AI semis market assumptions as of September 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable AI company type; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Incremental Semi Spend AI is the latest tech wave to inflect the semiconductor industry Models Cloud, DC, Semis Developer Tools Applications

• 76. 76 Source: Coatue opinion and analysis as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable AI company type; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. The entire semiconductor supply chain has potential to benefit Semis value chain across manufacturing inputs to device assembly IC design/embedded software Fabrication, pkg. & testing Materials Capital equip. Assembly Ex: Silicon wafer Ex: EUV Lithography Device assembly Semi design and manufacturing Semi manufacturing inputs End product: NVIDIA H100 Ex: NVIDIA chip design Ex: 3nm chip Data center servers powering AI! Models Cloud, DC, Semis Developer Tools Applications

• 77. Demand for cutting edge GPUs has exceeded supply 77 Source: Coatue analysis and opinion, as of November 2023. For illustrative purposes only; there is no guarantee that the demand for GPUs will continue to increase. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. H100 GPU prices on aftermarket are typically well above list price Pricing of individual H100 80GB GPU units on eBay & Amazon, not reflective of cluster availability Models Cloud, DC, Semis Developer Tools Applications $25,000 $30,000 $35,000 $40,000 $45,000 $50,000 05/23 06/23 07/23 08/23 09/23 10/23 11/23 List price Aftermarket price

• 78. Despite intense demand, AI compute costs have decreased 78 Could AI compute resemble cost curves in other sectors? Cost of running GPT-3/3.5 down 90% in 5 months! Source: OurWorldInData, Genome.gov, OpenAI pricing, Coatue analysis and opinion, as of November 2023. Price of Lithium batteries is per kWh, and cost of DNA sequencing is per megabase. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. 1990 1995 2000 2005 2010 2015 2020 $0 $3,000 $6,000 $9,000 $1,500 $4,500 $7,500 $10,500 -90% Price of lithium batteries Cost of DNA sequencing $0.02 $0.0020 0.000 0.005 0.010 0.015 0.020 0.025 Cost per 1K tokens ($) GPT-3 (Jan 2023) ChatGPT-3.5 (June 2023) -90% Models Cloud, DC, Semis Developer Tools Applications

• 79. Models on edge could alleviate GPU shortage as well 79 Source: ExLlama, MLC, OpenAI token generation tracking from GitHub, Nomic, Coatue analysis and opinion, as of November 2023. Performance tracking is from third parties and may not be representative of performance normally. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable AI company type; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Local models on Apple silicon becoming as fast as models running on GPUs Models Cloud, DC, Semis Developer Tools Applications Tokens per second (higher the better) of LLMs 15 35 35 19 24 GPT-4 GPT-3.5 Turbo LLAMA-2 70B GPT4All Code4All LLAMA-2 70B 10 Run on GPUs Run on Apple silicon ExLlama

• 80. Will we reach AI’s Holy Grail? 80 Source: Coatue analysis and opinion, as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable AI company type; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Largest, most expensive models are still most performant….for now Models Cloud, DC, Semis Developer Tools Applications Limited “intelligence” SOTA Quality Expensive H100s/A100s required Cheap Runs on Apple Silicon LLAMA-1 LLAMA-2 OLLAMA Holy Grail: SOTA quality while running locally! GPT-3.5 GGML Existing tradeoff between quality and compute costs… we need to break this paradigm!

• 81. 81 AI has started to create a new ecosystem of tools Source: Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. AI models AI apps End users AI developer tools Cloud platforms Data center / hardware / power AI developer tools

• 82. AI Ops is a new category of tools for AI engineers ML Ops: Focused on creating ML models AI Ops: Focused on operationalizing Generative AI Data labeling Data labeling & curation Model training & development Model inference Monitoring Model hub Fine tuning Model inference LLM Ops Vector DBs Monitoring & evaluation Illustrative; non-exhaustive Model hub Output Guardrails Nomos AI Source: Coatue analysis and opinion as of November 2023. For illustrative purposes only. Not exhaustive of all companies in these categories. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable AI company type; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Models Cloud, DC, Semis Developer Tools Applications Beyond labeling, data curation is becoming more important as engineers try to enhance model performance and alignment. New tools enabling fine tuning and inference have become easier for the AI developer to use 3 Emerging category of tools enabling developers to use LLMs more effectively; Vector DBs have exploded in popularity due to embeddings. Securing the last mile deployment of LLMs through model evaluation, guardrails, and ongoing RLHF is supporting a new set of companies 4 2 1

• 83. Curate Analyze Preparation Data curation tools will be critical for improving models 83 Source: Cohere as of November 2023, Coatue opinion and analysis as of September 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable AI company type; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Data curation an emerging new category between data collection & labeling, and model training Data collecting & labeling Dataset curation (New category) Model training Labeled data feeds ML models Automatically clean and scrub data for errors Profile the dataset across various metrics (e.g. token length distribution, entity label distribution) Filter data to include best examples for training; discard lowquality samples & identify human errors in labeling Pre-training Fine-tuning RLHF Curating datasets can reduce costs & improve quality. Training on ~30% of dataset can be sufficient! (from Cohere study) Models Cloud, DC, Semis Developer Tools Applications

• 84. Source: Hugging Face, Coatue opinion and analysis as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. 84 Fine tuning models has become much more accessible Hugging Face Autotrain lets you fine tune open-source models on your own data in just a few clicks Models Cloud, DC, Semis Developer Tools Applications 5 Select type of task & upload training dataset 1 2 3 Hugging Face suggests suitable models Select model(s) and evaluate model training progress

• 85. LLM Ops & Vector DBs are a new enabling layer for AI apps 85 Tools are already being put into production LLM Ops & Vector DBs enable retrieval augmentation Source: PyPI downloads, Coatue opinion and analysis as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable AI company type; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Cumulative installs (incl. machine) LLMs LLM Ops platforms Vector DBs AI Applications Unstructured data Embeddings model indexes unstructured data Embeddings stored in vector DBs; provides similarity search given query from LLM orchestrator LLM Ops chain and manage prompts + language model responses, and serves this to the end application Outputs response to user query, with context provided by LLM orchestrator Models Cloud, DC, Semis Developer Tools Applications 10M 20M 5M 15M 25M 30M 11/22 01/23 03/23 05/23 07/23 09/23 11/23

• 86. Graph DBs Real-time streaming DBs Vector DBs OLTP ~$25B mkt cap OLAP Can VectorDBs be standalone winners? 86 Only three DB types have significant public outcomes Startups have tried to create new categories of DBs Source: CapitalIQ, Coatue opinion and analysis as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable AI company type; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Observability ~$50B mkt cap No large standalone outcomes yet… … Will vector DBs be different? ~$25B mkt cap Models Cloud, DC, Semis Developer Tools Applications

• 87. General purpose databases are being used as VectorDBs too 87 Source: Supabase data as of June 2023, Coatue opinion and analysis as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See AppendixDisclosures for important disclosures, including regarding projections and forward-looking statements and trends. % of Supabase projects using pgvector Models Cloud, DC, Semis Developer Tools Applications 10% 2% 4% 6% 8% 10% 12% 01/23 02/23 03/23 04/23 05/23

• 88. Problem: Model evaluation is broken today 88 Humans prefer Claude 1 to Claude 2… …But Claude 2 performs better on benchmarks Source: LMSYS, Anthropic, Coatue analysis and opinion, as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable AI company type; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Models Cloud, DC, Semis Developer Tools Applications Overall win rates (human preference) against other models Performance on different benchmarks 77% 72% 0% 10% 20% 30% 40% 50% 60% 70% 80% Claude 1 Claude 2 73% 56% 85% 77% 71% 88% Bar exam Coding proficiency Grade school math Harmlessness Base 2x better Claude 1 Claude 2 LMSYS

• 89. Evaluation is one of many ‘last mile’ tools needed for AI in production 89 Last-mile tooling can improve AI accuracy and alignment This may help prevent hallucinations in critical sectors! Source: LinkedIn Post (Sarah Lukens 2023), Coatue opinion and analysis as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See AppendixDisclosures for important disclosures, including regarding projections and forward-looking statements and trends. Evaluate AI models Apps Iterate Fine-tuning and incontext learning Guardrails Feedback = Example last mile tooling Industrials case study: AI applied to prognostics & health mgmt. (PHM) 5% 10% 4% 29% Maintenance & Reliability Knowledge Exam PHM Knowledge Exam ChatGPT Bard % of questions with hallucinated answers Models Cloud, DC, Semis Developer Tools Applications

• 90. Deep Learning Era Rise of ML Engineers Era of AI Ops Natural language programming <100K ~300K ~30M Illustrative # of Addressable Developers ~8B Deep Learning Frameworks primarily usable only by ML researchers Transformers and ML Ops making building with models easier for more ML engineers We’re excited that AI Ops enables more developers to build apps 90 Source: IADSS as of September 2023, Tencent/TheVerge as of 2017, Coatue opinion and analysis as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable AI company type; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. New AI Ops tooling is accessible to all developers Example companies Example companies Example companies Pre-2015 2015-2022 2023+ As soon as 2024? = future outcomes = existing outcomes Example companies AI Ops tools enables all developers to be AI engineers Could OpenAI’s GPTs become the first way billions of people build AI apps with no code? Models Cloud, DC, Semis Developer Tools Applications GPTs

• 91. 91 The application layer is where humans will interact with AI Source: Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable AI company type; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. AI models AI apps End users AI developer tools Cloud platforms Data center / hardware / power AI apps

• 92. We expect AI to become even more impactful and autonomous Creativity Accuracy Copilots for skilled workers Creative content generation AI lawyer AI scientist AI doctor AI exec. AI skilled labor AI marketers AI influencers AI artists AI content creators Source: Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable AI company type; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Human + AI Copilot AI Autopilot AI Agent + Human Copilot 92 Models Cloud, DC, Semis Developer Tools Applications Today Future

• 93. We have seen this before: Copilot phase was short-lived in chess 93 Human peak Computer chess rating Computer chess rating 2005: Last win of computer-assisted human against machine 1997: IBM’s Deep Blue beats Kasparov Grandmasters Competitive Chess Players Machine Only “No doubt that a human paired with a computer is better than the single strongest computer in chess” – Gary Kasparov 1965 1975 1985 1995 2005 2015 1500 2000 2500 3000 3500 Human > Machine Human + Machine > Machine Machine beats all ELO Rating Coatue opinion and analysis as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Models Cloud, DC, Semis Developer Tools Applications

• 94. AI adoption accelerating across creative modalities 94 Usage of Runway’s AI video tools over time Midjourney Discord members over time Source: Runway usage as of June 2023 showing cumulative usage including video generation, AI rotoscoping, etc. Midjourney Discord members as of June 2023. Coatue opinion and analysis as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable AI company type; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forwardlooking statements and trends. Models Cloud, DC, Semis Developer Tools Applications 11/22 12/22 01/23 02/23 03/23 04/23 05/23 06/23 1M 2M 2M 3M 4M 8M 11M 16M +2,219% 11/22 12/22 01/23 02/23 03/23 04/23 05/23 06/23 4M 5M 7M 11M 13M 14M 15M 17M +342%

• 95. AI ramping up for design use cases in professional settings 95 Source: Aurora Solar and Tome as of June 2023, Coatue opinion and analysis as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Models Cloud, DC, Semis Developer Tools Applications Cumulative Tome signups 1M 2M 4M 6M 8M 10M 01/23 02/23 03/23 04/23 05/23 06/23 +1,191% AI solar designs generated 1,000 8,000 75,000 215,000 Q2 2022 Q3 2022 Q4 2022 Q1 2023 +21,400%

• 96. AI is inflecting within software development tools too 96 AI projects on Replit (annualized) Cumulative VSCode installs Source: Replit as of June 2023, Coatue opinion and analysis as of November 2023 (incl. Codeium installs). For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable AI company type; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Models Cloud, DC, Semis Developer Tools Applications Q1 ’22 Q2 ’22 Q3 ’22 Q4 ’22 Q1 ’23 Q2 ’23 3K 5K 8K 26K 123K 328K +12,042% 12/22 01/23 02/23 03/23 04/23 05/23 06/23 07/23 08/23 09/23 321K 3K 8K 19K 42K 75K 115K 159K 208K 262K +12,105%

• 97. Massive apps created on both sides of S-Curves of prev. cycles 97 Source: Capital IQ, World Bank as of September 2023, Statista as of September 2023, Coatue opinion and analysis as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable AI company type; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Application tech companies founded and US market cap along S-Curves 1990 2000 2010 2020 0% 1992 1994 1996 1998 2002 2004 2006 2008 2012 2014 2016 2018 2022 2024 20% 40% 60% 80% 100% Internet Smartphones 1994 – 1998 $3.2 Trillion 2002 – 2006 $900 Billion 2008 – 2011 $200 Billion 2012 – 2016 $250 Billion Models Cloud, DC, Semis Developer Tools Applications

• 98. 98 Incumbents gained most in smartphones: what about AI? Incumbents (Pre-2007) Native Mobile First (Post-2007) $6.5 Trillion $500 Billion $7 Trillion $500 Billion $500 Billion 2008 Mkt Cap Today’s Mkt Cap 12x!! Smartphone Era Gain Source: Capital IQ and Coatue analysis and opinion as of November 2023. For illustrative purposes only. This is showing solely gain in market cap after the smartphone era, and is not claiming causation of mobile on market cap. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable AI company type; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Models Cloud, DC, Semis Developer Tools Applications

• 99. 99 Models Cloud, DC, Semis Developer Tools Applications AI could drive incumbent price increases Microsoft Office Copilot could be an 80%+ increase in ARPU for Office users $14 $36 $36 $30 On-Prem* Cloud Cloud + AI $66 +157% +83% Per seat per month cost of Microsoft 365 E3 license Source: Microsoft pricing as of November 2023 (*On-Prem per month pricing takes Office 2010 Box Professional Pricing of $499, and assumes a 3 year usage. Note that Microsoft AI Copilot pricing of $30 per month is based on publicly announced pricing). Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends.

• 100. Is AI a game of kings or attackers? Incumbent Kings AI Native Attackers Fast moving incumbents > AI natives > Laggard incumbents Slow and risk averse Data Distribution & audience Full support infrastructure Large capital reserves Need for external funding Fast and nimble Creative and risk-on No tech debt Attract higher density of tech talent Source: Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Models Cloud, DC, Semis Developer Tools Applications 100

• 101. 101 The biggest mobile startups unlocked new behaviors… On Demand Access to Underutilized Assets +$215B Mkt Cap Real-time Photo Sharing & Messaging +$250B Mkt Cap Mobile Technology GPS Mobile Camera Mobile Internet App stores Mobile Fintech +$80B Mkt Cap Source: TikTok valuation taken from last Bytedance fundraise from G42 in Mar 2023. CapitalIQ and Coatue analysis and opinion as of November 2023. For illustrative purposes only. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable AI company type; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Models Cloud, DC, Semis Developer Tools Applications

• 102. AI startups should aim to create new behaviors too Artificial Intelligence “Most entrepreneurial ideas will sound crazy, stupid and uneconomic, and then they'll turn out to be right” —Reed Hastings Your invention here… AI autopilots? Models that can reason? Infinite gaming worlds? Source: Wired as of 2012, Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Models Cloud, DC, Semis Developer Tools Applications

• 103. Example: Character AI creating new virtual companions 103 Source: Coatue opinion and analysis as of November 2023. For illustrative purposes only. Highlighted company is an example of applicable AI company type; do not necessarily represent Coatue investments. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Users flocking to Character to chat with AI personas – brand new behavior! Models Cloud, DC, Semis Developer Tools Applications Monthly active web users Average time spent per visit 01/22 03/22 05/22 07/22 09/22 11/22 01/23 03/23 05/23 07/23 09/23 0M 2M 4M 6M 8M 10M 12M 14M 16M 18M 0 5 10 15 20 25 30 35 11/23

• 104. Example: AI as an autopilot could transform org structures 104 Previously, scaling businesses meant scaling headcount With AI, scaling business means scaling compute! Source: Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Models Cloud, DC, Semis Developer Tools Applications Product Sales & Support Operations Senior management Executive leadership Engineering Growth & Marketing Finance & Legal Product & Engineering AI autopilot Leadership Sales, support, and marketing AI autopilot Finance, legal AI autopilot

• 105. 105 There remains huge opportunity across modalities Category Example opportunities Open challenges Source: (1) Software defined by Statista as Productivity, Application Development, System Infrastructure, and Enterprise software. IFPI, Public Filings, Statista, Zion Market Research, Coatue opinion and analysis as of November 2023. For illustrative purposes only; does not represent an exhaustive list of opportunities or challenges. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Audio Video Code Search Other (e.g. robotics) AI musicians Text-to-video Software synthesis LLM-based search Robotic perception $600B+ size of global software1 market today $26B+ size of music industry today $162B+ Google Search revenue today $35B+ size of robotics market today $100B+ size of global movie & entertainment industry today • Hallucinations • Code security & robustness at scale • Integrations with existing SW deployment process • Costs to use agent-automated search • Response latency • Incumbent advantages in search data & distribution • Copyright infringement with artists • Incumbent distribution effects more pronounced (e.g. Apple audiobooks, Spotify) • Photorealism • Content moderation • Compute costs at scale • Hardware costs and upfront capital outlays • Unproven text-to-action workflow Models Cloud, DC, Semis Developer Tools Applications AI could increase the opportunity size of all these markets!

• 106. k IQaaS + Hardware Users ~$5T Market Cap Hardware + OS Mobile era AI era Users AI has potential to unlock “the super app of the West” Web browser Standalone Apps ~$2.5T Market Cap AI Super App one-stop-shop + actions Source: Capital IQ and Coatue analysis and opinion as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable AI company type; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Super Apps ~$600B Market Cap 106 Models Cloud, DC, Semis Developer Tools Applications

• 107. 107 Coatue view: the best of AI is yet to come Key Topics

• 108. Unimodality Multimodality Current gen Next gen Coatue View: The future top coding language will be English Source: Coatue opinion and analysis as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable AI company type; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. English Data Center Everywhere Research Engineering Public data Private data 108 Today Future Copilot systems that suggest code Foundation models targeted at code generation Natural language to programming, debugging, and deployment Ghostwriter Emerging coding agents Ghostwriter Factory From idea expressed in English… … To fully fledged application in minutes! Ghostwriter With help of future AI systems… Cursor Cursor And future companies?

• 109. Unimodality Multimodality Current gen Next gen English Data Center Everywhere Research Engineering Coatue View: On-device AI will become more widespread Source: MLC, Web Search for Apple AI strategy rumors, Coatue opinion and analysis as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable AI company type; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. 109 Public data Private data Future: Rumored Apple AI strategy Will everyone have a powerful LLM in their pocket? Apple Shortcuts Ajax GPT Siri + Today: MLC LLM allows you to run models on your phone + = 1B+ iPhone users become AI users 13 9 RedPajama 3B Vicuna 7B Tokens per second

• 110. Unimodality Multimodality Current gen Next gen English Data Center Everywhere Research Engineering Public data Private data Coatue View: Scaling AI is an engineering challenge Source: Coatue opinion and analysis as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Engineering headcount at OpenAI has increased over 3X since GPT-3 came out Estimated Y/Y growth rate in OpenAI engineering headcount 10% 20% 30% 40% 50% 60% 70% 80% 90% 01/2020 07/20 01/21 07/21 01/22 07/22 01/23 07/23 01/24 GPT-3 ~130 engineers Chat-GPT ~280 engineers Sep 2023 ~430 engineers Aug 2023 saw ~70% Y/Y engineering headcount growth! Estimated # of engineers 110

• 111. Public data Private data English Data Center Everywhere Research Engineering Coatue View: Private datasets can unlock new use cases Source: Coatue opinion and analysis as of November 2023. Companies and logos above are non-exhaustive and should only be construed as examples in a particular category. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. Logos listed above as examples of the applicable AI company type; do not necessarily represent Coatue investments. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. Unimodality Multimodality Current gen Next gen 111 Imagine what AI can do with Proprietary Data! Biotech Personalized medicine Media Generate shows ondemand 3D Models Engineer anything Gaming Infinite worlds & content eCommerce AI shopping concierge

• 112. English Data Center Everywhere Research Engineering Public data Private data Unimodality Multimodality Coatue View: Innovation in multimodality is a new frontier Source: Weaviate blog as of June 2023, Coatue opinion and analysis as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See AppendixDisclosures for important disclosures, including regarding projections and forward-looking statements and trends. Future: Lots of innovation within multimodal embeddings space! Today: GPT-4V is already creating new experiences for users + ChatGPT = Step by step written instructions if you get confused by IKEA diagrams! Open questions: Does training on multimodal datasets enable better reasoning capabilities for models? How can we keep multimodal models safe to use? What interesting emergent properties can we see from using the multimodal embeddings space? How do we scale multimodal datasets to train better models? Multimodal embeddings map relationships across text, image/video, and audio Current gen Next gen 112

• 113. English Data Center Everywhere Research Engineering Public data Private data Current gen Next gen Unimodality Multimodality Coatue View: New advancements in architecture coming Source: Stanford Hazy Research Lab, Arxiv, Coatue opinion and analysis as of November 2023. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See AppendixDisclosures for important disclosures, including regarding projections and forward-looking statements and trends. Cutting edge AI research emerging to improve Transformers Future | New architectures: Hyena Today | Architecture improvements: Flash Attention 25% 56% 72% Baseline Flash Attention (2022) Flash Attention 2 (2023) Model FLOPS utilization (i.e. efficiency) Could Hyena architecture enable million-token length context windows? 113

• 114. Enterprise deployment Compliance guardrails AI ethics & security AI authentication & permissioning Modalities Multi-modal models Embodied robotics Long-form video generation Availability Abundant hardware On-device AI & lowcost deployments SOTA models available for all Reducing Hallucinations Model output Discovery & novel insights Model explainability + more challenges yet to come… 114 Solving AI’s challenges has potential to unlock vast opportunities Generally useful AI for all Near term Longer term Source: Coatue opinion and analysis as of November 2023. List of challenges are not exhaustive and may change as AI landscape changes. For illustrative purposes only. There is no guarantee that Coatue’s views and projections regarding the future potential of AI are accurate or that any particular Coatue investment or fund will benefit from the AI trend. See Appendix-Disclosures for important disclosures, including regarding projections and forward-looking statements and trends. If you are a founder innovating in this space, we would like to connect with you. Please reach out at ai@coatue.com

• 115. 115 Appendix - Disclosures ¡ This whitepaper reflects Coatue’s opinions and analysis on AI as of the date of this presentation and does not constitute investment advice or a recommendation to buy or sell any securities. Since AI is an emerging technology, we expect our views may evolve or shift over time. As such, information herein is subject to change at anytime without notice. ¡ Unless otherwise indicated, any figures and calculations herein are for informational purposes only, computed by Coatue or its advisors and not audited by any third party. ¡ Although Coatue believes that the data expressed in this presentation is accurate and reasonable, actual results could differ materially from those projected or assumed, and such projections are subject to change, and are subject to inherent risks and uncertainties. Neither Coatue nor its affiliates or advisors guarantees the accuracy or completeness of the information. Accordingly, neither Coatue nor any of its affiliates, advisors or employees shall be liable to you or anyone else for any loss or damages from use of the information contained in this presentation. Coatue Analysis Forward-looking Statements & Projections ¡ The companies referenced herein are included for informational purposes only. The information herein does not constitute investment advice or a recommendation to buy or sell any securities. The companies do not necessarily represent stocks or investments that Coatue owned or owns. In addition, the trends identified and discussed in this presentation reflect the opinions of Coatue. The trends discussed do not reflect the entire universe and could be impacted by market factors, changes in laws and other factors. ¡ No third-party firm or company names, brands or logos used in this presentation are Coatue’s trademarks or registered trademarks, and they remain the property of their respective holders and not Coatue. The inclusion of any third-party firm and/or company names, brands and/or logos does not imply any affiliation with these firms or companies. None of these firms or companies has endorsed the investment opportunity described herein, Coatue, any affiliates of Coatue, or Coatue’s personnel. Companies and Trends ¡ This presentation contains forecasts, projections and other forward-looking statements, including (but not limited to) the occurrence or outcome of anticipated events, estimates, future performance and adaption of AI. Due to various risks and uncertainties, actual events, results of these events may differ materially from those reflected or contemplated in such forward-looking statements. There is no guarantee that such forecasts, projections or forward-looking statements will occur and therefore should not be relied upon. Artificial Intelligence (AI) ¡ This presentation contains forward-looking predictions regarding AI and its potential impacts and opportunities, all of which are subject to a number of factors and uncertainties. Any characterization of AI herein is the opinion of Coatue, is subject to change, and should not be relied upon in making an investment decision. Given that AI is an emerging technology, assessing the future trajectory of the AI industry is inherently challenging, and Coatue’s views on its success or failure can be subjective and based on incomplete information, limited perspectives, or speculative assumptions. See also the disclosures regarding forward-looking statements. ¡ Companies herein are not intended to highlight or represent the Coatue portfolio, but rather the broader AI theme, which by nature may include Coatue investments. To the extent Coatue portfolio companies or investments are included herein, Coatue makes no suggestion or guarantee regarding the future outcomes or performance of such companies. ¡ Even if Coatue’s characterizations and opinions regarding the AI trend were to prove accurate, there is no suggestion or guarantee that Coatue will be able to identify and invest in opportunities presented by AI. The AI industry is multifaceted, encompassing various technologies, applications, and market dynamics. Its complexity makes it susceptible to unpredictable developments, including breakthrough innovations, disruptive technologies, or unexpected challenges.