Jonathan Gorard

@getjonwithit

Co-founder and CEO @lanyon_ai, making the universe computable. Part-time math/physics research @Princeton Previously @Cambridge_Uni @WolframResearch

following
followers

What could compensate if mechanistic interpretability is infeasible?

A user argues computational irreducibility makes mechanistic interpretability infeasible, drawing parallels with fluid mechanics and psychology.

1 Source, 6d ago, first seen 6d ago

The value of understanding a hypothetical cancer cure's mechanism

A poster says almost any Stage 4 cancer patient would find human understanding of a working cure's mechanism irrelevant.

20 Sources, 12d ago, updated 12d ago

OpenAI releases results from its internal AI math model on GitHub

OpenAI says it drew on an independent mathematics advisory group’s advice about how to release the results.

435 Sources, 12d ago, updated 12d ago

Can physics get a proof checker?

A post introducing a conversation with applied mathematician Jonathan Gorard says it explores why AI took off in math and coding, but not physics.

1 Source, 17d ago, first seen 17d ago

Three meanings of “computer” in the brain-as-computer debate

A commenter says the brain is not a von Neumann machine, but whether broader computer models suffice remains an open question.

1 Source, 18d ago, first seen 18d ago

Vibecheck hackathon to focus on building and formally verifying software

Organizers welcome solo builders and teams; a sponsor promises all participants early access to Lanyon AI tools and unlimited compute.

1 Source, 22d ago, first seen 22d ago

The hurdles to AI breakthroughs in physics

One post argues that major physics breakthroughs require insights beyond current AI's reach. Another points to a lack of formal verification in physics and claims Lanyon AI is closing that gap.

3 Sources, 26d ago, updated 26d ago

A call for e/acc to keep accelerating toward Kardashev 3

One post declares “e/acc is forever” and urges continued movement “up and to the right” on the Kardashev scale.

16 Sources, 19d ago, updated 19d ago

“This inference could have been a tool call,” a user quips

A user’s one-line AI joke frames a tool call as an alternative to an inference.

1 Source, 41d ago, first seen 41d ago

Alexander Terenin Calls AI a Mirror to Human Hierarchies

Cornell AI researcher posts quote comparing AI to social media as a revealing mirror.

15 Sources, 49d ago, first seen 49d ago

Lanyon AI CEO Gives Princeton Talk on Formal Verification

Lanyon AI shares details of its CEO's March 2025 Princeton talk on formal verification.

3 Sources, 50d ago, first seen 50d ago

Featured in

What could compensate if mechanistic interpretability is infeasible?

The value of understanding a hypothetical cancer cure's mechanism

OpenAI releases results from its internal AI math model on GitHub

Can physics get a proof checker?

Three meanings of “computer” in the brain-as-computer debate

Vibecheck hackathon to focus on building and formally verifying software

The hurdles to AI breakthroughs in physics

A call for e/acc to keep accelerating toward Kardashev 3

“This inference could have been a tool call,” a user quips

Alexander Terenin Calls AI a Mirror to Human Hierarchies

Lanyon AI CEO Gives Princeton Talk on Formal Verification

Featured in

What could compensate if mechanistic interpretability is infeasible?

The value of understanding a hypothetical cancer cure's mechanism

OpenAI releases results from its internal AI math model on GitHub

Can physics get a proof checker?

Three meanings of “computer” in the brain-as-computer debate

Vibecheck hackathon to focus on building and formally verifying software

The hurdles to AI breakthroughs in physics

A call for e/acc to keep accelerating toward Kardashev 3

“This inference could have been a tool call,” a user quips

Alexander Terenin Calls AI a Mirror to Human Hierarchies

Lanyon AI CEO Gives Princeton Talk on Formal Verification