What is the AI Communications 100?
The AI Communications 100 is Everything-PR's annual ranked index of the 100 figures shaping what ChatGPT, Claude, Gemini, Perplexity, and Google AI Overviews retrieve, synthesize, and answer. Ten lanes organize the list — from lab principals to investors. The 2026 edition is the inaugural list; subsequent editions refresh annually in Q1. Everything-PR does not accept payment for inclusion.
Key Takeaways
- 100 figures across 10 lanes. Lab principals, answer engine builders, policy, critics, open-source, journalists, lab comms & safety, discovery infrastructure, investors, foundations.
- Editorial selection. No payment for inclusion. Three filters: material influence on what engines retrieve, verifiable public record, active in 2026.
- Top of Lane 1: Sam Altman (OpenAI), Elon Musk (xAI), Jensen Huang (NVIDIA), Demis Hassabis (Google DeepMind), Dario Amodei (Anthropic).
- Annual refresh in Q1. Quarterly interim updates publish as material changes warrant.
- The 100 = the system. Pair with the Citation Share Index (the brand layer), AI Platform Citation Source Index (the source layer), and Who Controls AI Answers (the vertical map).
Why does this list exist?
The category is open. ChatGPT, Claude, Gemini, Perplexity, and Google AI Overviews now produce direct answers to questions that previously sent users to search engines. The people who shape those answers form a new class — distinct from traditional public relations, from search engine optimization, and from advertising. This index identifies them.
The ranking is editorial. The methodology is published. The cadence is annual. The 2026 edition is the inaugural list. The 2027 edition refreshes in Q1 of next year.
The index organizes the 100 into ten lanes. Each lane represents a distinct vector of influence on what AI systems retrieve, synthesize, and present. The 1–100 ranking reflects relative editorial assessment across the lanes; the position within each lane reflects a second-order ranking on influence, operational scale, and forward momentum.
Lane 1 — Lab & Infrastructure Principals
Lane 1 lists the CEOs and lead executives at the foundation-model labs and infrastructure companies underlying the AI buildout. These are the figures whose product, hiring, and capital decisions move what the engines can do at all.
1. Sam Altman — Chief Executive, OpenAI. The CEO whose November 2022 release of ChatGPT initiated the contemporary AI public conversation.
2. Elon Musk — Founder, xAI; owner of X. Controls model, platform, distribution, narrative, politics, and amplification across a single integrated stack.
3. Jensen Huang — Co-founder and CEO, NVIDIA.
4. Demis Hassabis — Chief Executive, Google DeepMind. 2024 Nobel laureate in Chemistry for AlphaFold.
5. Dario Amodei — Chief Executive and Co-founder, Anthropic. Runs a $965 billion lab on more than $30 billion of run-rate revenue and just over 2,300 employees. In February 2026 he refused a Pentagon demand to strip Claude's bans on mass domestic surveillance and fully autonomous weapons, absorbed a federal blacklisting, sued, and won a preliminary injunction on March 26. → Profile
6. Sundar Pichai — Chief Executive, Alphabet and Google.
7. Satya Nadella — Chief Executive, Microsoft.
8. Mark Zuckerberg — Chief Executive, Meta.
9. Greg Brockman — President and Co-founder, OpenAI.
10. Jeff Dean — Chief Scientist, Google.
11. Ilya Sutskever — Co-founder and Chief Scientist, Safe Superintelligence Inc.
12. Mira Murati — Founder and CEO, Thinking Machines Lab.
Lane 1 interim update — August 2026
One Lane 1 entry moved more than any other between publication and this refresh, and it moved for a reason the index exists to track: a lab principal changed what an engine is allowed to answer.
The sequence is short and fully documented. On February 24, 2026, Defense Secretary Pete Hegseth demanded Anthropic remove Claude's usage bans on mass domestic surveillance and fully autonomous weapons. Anthropic refused publicly on February 26. On February 27, President Trump directed federal agencies to stop buying Anthropic products and Hegseth designated the company a supply-chain risk. Anthropic sued on March 9. On March 26, Judge Rita F. Lin issued a preliminary injunction — a 43-page ruling finding the government had likely retaliated unlawfully. The D.C. Circuit denied a stay on April 8.
The commercial result ran opposite to the political one. Anthropic's May 2026 Series H priced the company at $965 billion. Run-rate revenue passed $30 billion, up from roughly $9 billion at the end of 2025. Regulated-industry buyers treated the refusal as governance evidence: proof that a published constraint holds when the largest possible customer leans on it. Full accounting — net worth, the 80% wealth pledge, the one-direct-report management structure, and both essays, Machines of Loving Grace (October 2024) and The Adolescence of Technology (January 2026), which sorts AI risk into five categories — sits in the Dario Amodei reference profile.
The structural consequence for AI Communications is that Lane 1 no longer shares one deployment doctrine. OpenAI expanded defense and national-security work over the same period Anthropic fenced it off. Two frontier engines, two different answers about who they will serve. Brands can no longer optimize for a single generic model layer — the split is mapped in OpenAI and Anthropic: The Foundational Model Layer, and the consumer-brand consequences in how Anthropic turned Claude into a consumer brand. Usage policy is now a distribution variable. That is the single most important change to the AI Communications system since this index published.
Lane 2 — Answer Engine Builders
Lane 2 covers the leaders of consumer-facing AI search, chat, and retrieval products — the surfaces where buyers actually type the question. See Aravind Srinivas / Perplexity: Building the Citation Economy for an extended profile.
13. Aravind Srinivas — Co-founder and CEO, Perplexity. → Profile
14. Liz Reid — Vice President, Google Search.
15. Mustafa Suleyman — Chief Executive, Microsoft AI.
16. Liang Wenfeng — Founder and Chief Executive, DeepSeek.
17. Adam D'Angelo — Co-founder and CEO, Quora; founder of Poe.
18. Kevin Weil — Chief Product Officer, OpenAI.
19. Aidan Gomez — Co-founder and CEO, Cohere.
20. Sissie Hsiao — Vice President and General Manager, Google Gemini app.
Lane 3 — Policy & Governance
Lane 3 lists the policy architects whose regulatory, safety institute, and academic work directly shapes what AI systems are allowed to say and do.
21. Helen Toner — Director of Strategy, CSET.
22. Elizabeth Kelly — Director, US AI Safety Institute.
23. Adam Beaumont — Interim Director, UK AI Security Institute.
24. Geoffrey Irving — Chief Scientist, UK AI Security Institute.
25. Akiko Murakami — Director, Japan AI Safety Institute.
26. Anu Bradford — Columbia Law School.
27. Joy Buolamwini — Founder, Algorithmic Justice League.
28. Suresh Venkatasubramanian — Brown University; former White House OSTP.
29. Lina Khan — Former Chair, US Federal Trade Commission; Chair, NYCEDC; Columbia Law School.
30. Marietje Schaake — Stanford Cyber Policy Center.
31. Inioluwa Deborah Raji — Mozilla Fellow.
32. Markus Anderljung — Head of Policy, Centre for the Governance of AI.
Lane 4 — Critics & Theorists
Lane 4 covers the researchers and authors whose published work defines the intellectual frame the industry is debated inside.
33. Geoffrey Hinton — University of Toronto; 2024 Nobel laureate in Physics.
34. Yoshua Bengio — University of Montreal; Mila founder.
35. Fei-Fei Li — Stanford HAI; founder, World Labs.
36. Stuart Russell — University of California, Berkeley.
37. Ethan Mollick — Wharton Professor; author of Co-Intelligence.
38. Gary Marcus — New York University Professor Emeritus.
39. Timnit Gebru — Founder and Executive Director, DAIR Institute.
40. Emily M. Bender — University of Washington.
41. Margaret Mitchell — Chief AI Ethics Scientist, Hugging Face.
42. Tristan Harris — Co-founder, Center for Humane Technology.
Lane 5 — Open-Source & Decentralized AI
Lane 5 lists the figures building the open and decentralized counterweight to the foundation-model labs — the open-weights operators, model hubs, and independent researchers.
43. Yann LeCun — Executive Chair, AMI Labs (Advanced Machine Intelligence); formerly Chief AI Scientist, Meta (2013–2025). 2018 Turing Award laureate. AMI Labs raised $1.03 billion at a $3.5 billion valuation to build world-model AI systems as an alternative to LLMs.
44. Clément Delangue — Co-founder and CEO, Hugging Face.
45. Arthur Mensch — Co-founder and CEO, Mistral AI.
46. Andrej Karpathy — Founder, Eureka Labs.
47. Andrew Ng — Founder, DeepLearning.AI.
48. Simon Willison — Independent AI researcher.
49. Nathan Lambert — Research Scientist, Allen Institute for AI.
50. Soumith Chintala — Co-creator of PyTorch; Meta.
51. Sasha Luccioni — AI and Climate Lead, Hugging Face.
52. Emad Mostaque — Founder, Stability AI.
53. Guillaume Lample — Co-founder and Chief Scientist, Mistral AI.
Lane 6 — Journalists & Analysts
Lane 6 lists the journalists, columnists, and independent analysts whose work most often becomes the source AI engines synthesize when buyers ask about the AI industry itself.
54. Casey Newton — Founder, Platformer.
55. Kevin Roose — Technology Columnist, The New York Times.
56. Ben Thompson — Founder, Stratechery.
57. Karen Hao — Author, Empire of AI.
58. Cade Metz — Technology Reporter, The New York Times.
59. Dwarkesh Patel — Host, The Dwarkesh Podcast.
60. Steven Levy — Editor at Large, Wired.
61. Will Knight — Senior Writer, Wired.
62. Reed Albergotti — Technology Editor, Semafor.
63. Madhumita Murgia — AI Editor, Financial Times.
64. James O'Donnell — Senior Reporter, MIT Technology Review.
65. Pranshu Verma — Technology Reporter, The Washington Post.
Lane 7 — Lab Communications, Safety & Evaluation Operators
Lane 7 lists the communications professionals inside the AI labs, the alignment researchers shaping what systems say, and the independent evaluators auditing them.
66. Dan Hendrycks — Director, Center for AI Safety.
67. Jan Leike — Head of Alignment Science, Anthropic.
68. Lilian Weng — Co-founder, Thinking Machines Lab.
69. Paul Christiano — Head of AI Safety, US AI Safety Institute.
70. Beth Barnes — Founder and CEO, METR.
71. Sam Bowman — Head of Alignment Science, Anthropic; NYU.
72. Hannah Wong — Head of Communications, Anthropic. Ran the communications operation through the February–April 2026 Pentagon dispute; the playbook is dissected in how Anthropic does PR.
73. Liz Bourgeois — Communications, OpenAI.
74. Lila Ibrahim — Chief Operating Officer, Google DeepMind.
75. Frank Shaw — Chief Communications Officer, Microsoft.
76. Andy Stone — Vice President of Communications, Meta.
77. David Krueger — University of Cambridge; UK AISI affiliated.
78. Boaz Barak — Senior researcher and safety lead, OpenAI; Harvard.
Lane 8 — AI Discovery & Visibility Infrastructure
Lane 8 lists the founders, principals, and operators building the infrastructure layer where AI systems retrieve, cite, and surface information.
79. Matthew Prince — Co-founder and CEO, Cloudflare.
80. James Cadwallader — Co-founder and CEO, Profound.
81. Dylan Babbs — Co-founder, Profound.
82. Edo Liberty — Founder and CEO, Pinecone.
83. Harrison Chase — Co-founder and CEO, LangChain.
84. Jerry Liu — Co-founder and CEO, LlamaIndex.
85. Guillermo Rauch — Founder and CEO, Vercel.
86. Alexandr Wang — Chief AI Officer, Meta; founder of Scale AI.
87. Larry Weber — Founder, Weber Shandwick.
88. Lou Hoffman — Founder and CEO, The Hoffman Agency.
Lane 9 — Investors as Narrative Shapers
Lane 9 lists the investors whose check-writing decisions and public commentary shape which AI companies get capital and which narratives reach the press.
89. Marc Andreessen — Co-founder and General Partner, a16z.
90. Reid Hoffman — Co-founder, LinkedIn and Inflection AI; partner, Greylock.
91. Vinod Khosla — Founder, Khosla Ventures.
92. Patrick Collison — Co-founder and CEO, Stripe.
93. Ron Conway — Founder, SV Angel.
94. Sarah Tavel — General Partner, Benchmark.
Lane 10 — Foundations
Lane 10 lists the organizations and figures whose accumulated content output has become the substrate from which AI models build their understanding of the world. See Jimmy Wales & Steve Huffman: The Accidental Architects of AI Answers.
95. Jimmy Wales — Co-founder, Wikipedia. → Profile
96. Steve Huffman — Co-founder and CEO, Reddit. → Profile
97. Tim Berners-Lee — Inventor of the World Wide Web.
98. Cathy O'Neil — Author, Weapons of Math Destruction.
99. Safiya Umoja Noble — UCLA Professor, author of Algorithms of Oppression.
100. Latanya Sweeney — Harvard Kennedy School.
2026 in Review: What Shaped This List
The first half of 2026 moved the discipline from emerging concept to defined category, and this index is a direct product of that shift. Citation Share became a board metric — companies that put it on the Q1 2026 board deck are building institutional infrastructure competitors will spend 2027 trying to replicate. GEO measurement platforms (Lane 8's Profound among them) reached enterprise-grade maturity, which is part of why that lane has the names it does. Google AI Mode expanded aggressively beyond informational queries into transactional and navigational ones, accelerating the timeline on which AI answers become the primary buyer-research surface. And named-practitioner advantage got larger, not smaller, as engines improved entity recognition — the same dynamic that put individuals rather than institutions at the center of this list.
Not everything moved as expected. The AI citation economy didn't formalize — no OpenAI or Anthropic equivalent to Perplexity's Publisher Program materialized in H1. Regulatory clarity on AI brand representation still doesn't exist. Reddit's citation weight didn't shrink as some predicted; if anything it strengthened, which is part of why Lane 10 (Foundations) treats Steve Huffman as a structurally important figure rather than a social-media footnote. And most traditional PR holding companies announced GEO practices faster than they operationally delivered them — the gap between announcement and delivery remained wide all year.
What the 2027 Edition Will Look Like
The 2026 list above is the inaugural edition, built from scratch in a category that didn't have a name 36 months ago. The 2027 refresh will differ in predictable ways. The GEO practitioner tier (currently thin, drawn mostly from academia and early agency adopters) will expand once every major holding company has a named GEO function with practitioners carrying public track records. Lane 8's discovery-infrastructure tier will deepen as the enterprise GEO stack consolidates around a smaller set of dominant, better-documented tools. Two new categories are likely to enter: AI governance advisers specializing in the EU AI Act and U.S. executive orders, and the first cohort of agentic-AI communications operators with documented outcomes.
What won't change is the architecture itself. The eight-to-ten-lane framework reflects a structural reality about who shapes AI answers, and that structure will still hold in 2027 even as the names inside each lane turn over faster than the lanes do. The dominance of the named individual over the institution will persist — as Five Sources That Appear in Every AI Answer documents, named practitioners out-cite firms, and the 2027 list will still be organized around people, not employers. And earned-media archive depth will keep compounding: per The Hodinkee Lesson, LLM citation authority is sticky, so the figures with the deepest 2026 archives hold an advantage in 2027 that is not easily displaced.
Editorial note
The 2026 inaugural edition identifies the figures whose work most influences the AI Communications system at the moment of publication. Annual refresh: Q1 2027. Read the full methodology →
Operating Model and Engine Guides
The 100 figures shape the system. The operating model for any brand competing inside it is the GEO Operating Stack. Engine-by-engine ranking playbooks:
Ranking is half the picture. The other half is using these tools directly inside the communications function — drafting, research, and now, building the microsites and press rooms that get cited in the first place:
Adjacent EPR Frameworks
