Library intelligence

Leaders Watch

Tracking how influential figures' positions evolve in public—through statements, interviews, governance moves, and the documents we host. This is not a directory; it is an editorial timeline you can extend entry by entry. Use theme and role filters to slice the set (e.g. governance, alignment).

How we assign tiers

Tiers reflect present-day leverage over frontier AI and the global safety debate—who can move capability, deployment, norms, or regulation at the largest scale right now, not lifetime achievement alone. Evidence still has to be anchored to primary sources on each profile; tier is not a quality score for every quote.

  • Tier 1 (featured) — Heads of major frontier labs or hyperscaler AI divisions; founders of standalone frontier-model companies; and a small set of researchers whose public statements routinely reset institutional and media language on catastrophic or systemic risk. If their narrative or product choices shift, many others have to react.
  • Tier 2 — Highly influential specialists: alignment and safety researchers, norm-shaping advocates, and leaders whose AI footprint is significant but narrower than the tier-1 bar above. Still first-class profiles with full timelines—not a "B list" for rigour.
16 profilesTier-1 featuredTheme + role filters

Filter

Matching leaders

16 profiles match your filters.

Researcher

Andrew Chi-Chih Yao

Tsinghua IIIS leadership

Tier 1

Founding figure in computational complexity and cryptography whose later career focuses on elite computer-science education, interdisciplinary institutes, and AI-related research leadership in China—representing the “deep bench” beneath headline-grabbing models.

chinaresearchgovernanceinstitutionsrisk

Open timeline →

Founder / builder

Dario Amodei

Anthropic leadership

Tier 1

Researcher-operator whose public writing and interviews often stress interpretability, misuse, and deployment discipline as Anthropic scales competing models.

alignmentgovernancedeploymentriskalignment

Open timeline →

Executive

Demis Hassabis

Google DeepMind leadership

Tier 1

Research leader turned lab chief whose public mix of scientific optimism and cautious governance language tracks how DeepMind interfaces with Alphabet and global regulators.

governanceriskresearchinstitutionscapability

Open timeline →

Founder / builder

Elon Musk

xAI leadership

Tier 1

A high-decibel operator whose bets span frontier models (xAI), autonomy, and global platforms—continuously shaping how the public and policymakers imagine both catastrophic risk and competitive acceleration.

riskgovernancedeploymentcapabilityinstitutions

Open timeline →

Researcher

Geoffrey Hinton

Independent research leadership

Tier 1

A central figure in modern neural networks whose public risk framing in 2023 helped legitimise extinction-risk and governance language well beyond specialist circles.

riskgovernancepublic understandingalignmentgovernance

Open timeline →

Founder / builder

Kai-Fu Lee

01.AI / Sinovation leadership

Tier 1

A bridge figure between Silicon Valley and China’s AI industry whose public writing and investing track how capability, capital, and state-market dynamics intersect in the PRC—and how that compares to US frontier labs.

chinainvestmentgovernancesocietyrisk

Open timeline →

Executive

Mark Zuckerberg

Meta leadership

Tier 1

Operator of one of the world’s largest AI research and deployment footprints; the Llama family and open-weights strategy materially reset the debate over capability diffusion, misuse, and guardrails.

deploymentgovernanceriskopen-sourceinstitutions

Open timeline →

Executive

Mustafa Suleyman

Microsoft AI leadership

Tier 1

Product-facing executive for Microsoft’s consumer and copilot AI stack; public voice on containment, “frontier” safeguards, and how enterprise-scale deployment should pair with governance.

governancedeploymentriskinstitutionspublic understanding

Open timeline →

Executive

Robin Li

Baidu leadership

Tier 1

Operator of China’s largest search-and-AI platform stack; Baidu’s ERNIE models and cloud narrative are a primary domestic counterpart to Western foundation-model labs for governance and deployment analysis.

chinadeploymentgovernanceriskinstitutions

Open timeline →

Executive

Sam Altman

OpenAI leadership

Tier 1

A defining operator of the frontier-lab era whose public commitments on safety, deployment, and geopolitical context are continuously stress-tested by product velocity.

governanceriskdeploymentinstitutionscapability

Open timeline →

Researcher

Yoshua Bengio

Mila / academic leadership

Tier 1

A co-founder of modern deep learning who has moved into explicit AI safety and governance work, including institutional experiments that bridge research and policy.

alignmentgovernanceriskinstitutionsgovernance

Open timeline →

Researcher

Ilya Sutskever

Safe Superintelligence leadership

A core technical architect of the deep learning boom whose later public emphasis tilted toward superintelligence risk and governance—often in tension with product tempo.

alignmentgovernanceriskalignmentgovernance

Open timeline →

Advocate / organiser

Max Tegmark

FLI leadership

A public organiser and communicator who links existential risk framing to concrete policy instruments and institutional coordination.

governanceriskinstitutionssocietygovernance

Open timeline →

Researcher

Paul Christiano

Alignment research leadership

A leading technical voice on scalable oversight, iterated amplification, and practical alignment paths—often bridging abstract risk to engineering agendas.

alignmenttechnicalgovernancealignmentevals

Open timeline →

Researcher

Stuart Russell

Academic governance leadership

Longstanding public interpreter of AI risk for policymakers and general audiences—often with formal models of uncertainty and control.

governanceriskalignmentpolicyalignment

Open timeline →

Advocate / organiser

Tristan Harris

Center for Humane Technology leadership

Former design ethicist turned organiser focused on attention economies, platform power, and how AI accelerates psychological and civic harms.

societygovernancedemocracyriskdemocracy

Open timeline →