New York Magazine’s Intelligencer section has released its annual 'Power' issue, focusing on the future of AI. The article, titled 'Powerful People Shaping the Future of AI,' identifies 26 individuals whose influence far exceeds their public fame. Rather than just listing CEOs, the piece highlights engineers, researchers, and product leaders who are actively defining the technical and operational boundaries of modern AI systems. The selection process involved collaboration with the publication Transformer and interviews with over 50 industry leaders to map the subtle power dynamics within the sector.

The Rise of the Niche Expert

The report emphasizes that AI power is increasingly concentrated in specialized technical roles rather than traditional corporate hierarchies. Logan Graham, who leads Anthropic’s Frontier Red Team, exemplifies this shift. Graham’s work involves stress-testing models for catastrophic failures, such as the April 7 announcement that Claude Mythos Preview surpassed standard cybersecurity benchmarks. By identifying that the model could exploit security bugs in major operating systems, Graham’s team directly influenced Anthropic’s decision to withhold the model from the public while granting limited access to select partners like AWS and Microsoft. This case demonstrates how a single researcher’s findings can dictate product rollout strategies for billions of dollars in infrastructure.

Infrastructure and Insight

Dylan Patel, founder of SemiAnalysis, is highlighted as a critical influencer in hardware and infrastructure. Patel’s firm provides deep technical analysis of chip supply chains and data center economics, effectively becoming a required resource for investors and executives. Jensen Huang, CEO of Nvidia, explicitly cited SemiAnalysis as a validator of Nvidia’s chip quality, underscoring Patel’s industry weight. SemiAnalysis is projected to generate $100 million in revenue this year and has raised a $500 million fund, blurring the lines between media analysis and investment strategy. This convergence allows Patel to shape market narratives around hardware bottlenecks and software limitations, such as his critique of AMD’s software stack, which prompted a direct response from CEO Lisa Su.

Product-Led Growth in AI

Thibault 'Tibo' Sottiaux, a senior executive at OpenAI, represents the shift toward accessible AI products. Sottiaux pivoted the Codex team from a command-line interface to a native desktop application, recognizing that complex agent interactions required a more intuitive user experience. This strategic move resulted in Codex reaching 2 million weekly users by March 19, leading to Sottiaux’s promotion to head of core products. His approach transformed Codex from a niche developer tool into the foundation of ChatGPT’s desktop experience, proving that user interface design is now a primary driver of adoption in the AI space. The article notes that such product-led decisions are reshaping how non-technical users interact with AI agents.

Key Takeaways

  • Influence in AI is increasingly held by specialized engineers and analysts rather than just corporate executives.
  • Red teaming efforts, such as those led by Logan Graham at Anthropic, directly impact public access to frontier models.
  • Infrastructure analysts like Dylan Patel are shaping hardware investment trends and vendor strategies through technical deep dives.
  • Product pivots, such as Thibault Sottiaux’s shift of Codex to a desktop app, are critical for scaling AI adoption beyond developer communities.

The Bottom Line

For developers and builders, this list confirms that technical depth and product intuition are the new currencies of power in the AI industry. Success now depends on mastering specific domains like security, infrastructure, or UX, rather than climbing traditional corporate ladders.