A field guide to AI positions

Ilya Sutskever

Sources used: First-hand material.

The map position is our interpretation of the public statements below. Read the method.

Position, range and evidence details

Preferred pace: 15; stated catastrophic-risk concern: 72. These are layout units, not percentages or rankings.

Pace range: -35 to 65; concern range: 45 to 95. Ranges show alternative editorial interpretations, not statistical confidence intervals.

Evidence clarity: medium. Basis: primary. Implementation: public-advocacy.

Open on the map Reading notes · Structured record

What this position covers

Personal research ambitions and proposed limits on the most powerful superintelligence. [1] [2] [3]

Reference this explanation or suggest a correction

Link to this explanation · Suggest a correction · How corrections work

What to keep in mind

The proposed power cap has no specified method. Safety is a research aim, not an established property of future systems. [1] [2] [3]

Reference this explanation or suggest a correction

Link to this explanation · Suggest a correction · How corrections work

What they say about the pace of AI

Says his research is ready to be scaled with a larger computer. [1]

Find the passage in the source

Sutskever statement in the partnership announcement

Reference this explanation or suggest a correction

Link to this explanation · Suggest a correction · How corrections work

What they say about the pace of AI

Supports a cap on the most powerful superintelligence, while leaving the method open. [2]

Find the passage in the source

01:02:37 to 01:04:10

Reference this explanation or suggest a correction

Link to this explanation · Suggest a correction · How corrections work

What they say about catastrophic risk

His coauthored article warns of human disempowerment or extinction from superintelligence. [3]

Find the passage in the source

Opening and alignment problem

Reference this explanation or suggest a correction

Link to this explanation · Suggest a correction · How corrections work

What they say about catastrophic risk

Still treats extreme system power and control as central safety concerns. [2]

Find the passage in the source

00:56:10 to 01:04:10

Reference this explanation or suggest a correction

Link to this explanation · Suggest a correction · How corrections work

The wider picture

Pursues more capable systems while supporting limits on extreme power; control remains a research problem. [1] [2] [3]

Reference this explanation or suggest a correction

Link to this explanation · Suggest a correction · How corrections work

Research pace and release plans

In his November 2025 interview, he gives more weight to incremental release. That concerns deployment, not a general halt to capability research. [2]

Reference this explanation or suggest a correction

Link to this explanation · Suggest a correction · How corrections work

Share this page

https://theaiatlas.org/people/sutskever/

Download a share image · Vector image

Image previews are summaries. Keep the page link so readers can check the evidence.

Sources and what we read

  1. 1. Ilya Sutskever’s Safe Superintelligence Inc. and NVIDIA Announce Long-Term Strategic Partnership

    Read the issuer announcement linked from SSI, including Sutskever’s attributed research-scaling statement. Primary company announcement, not independent reporting or a capability test. SSI’s update is dated July 26; this release is dated July 27.

  2. 2. Ilya Sutskever — We're moving from the age of scaling to the age of research

    Read the host transcript, especially 00:43:04 to 00:46:47 and 00:56:10 to 01:04:10. The host describes editorially reworked transcripts. The proposed power cap lacks an implementation method; gradual release concerns deployment.

  3. 3. Introducing Superalignment

    Read the coauthored introduction, research approach and footnotes. Used for historically expressed extinction concern and the distinction between a research aim and a solved control problem. The announced timetable is not treated as a result.

Edition and machine-readable evidence

Content version 0.19.0. Evidence cutoff 2026-09-15; this does not mean every source was read on that day.

Pinned complete dataset · Complete evidence page · Agent consumption guide