cj

Automating ML accelerator design @itsasecret

[cj @cjrohandj]

@FloatHeadPhys I was just wanting a refresher for my device physics class! Great timing

[cj @cjrohandj]

@reelabduction @_anishlk Ratio

[anish lakkapragada @_anishlk]

@reelabduction holy have i finally made it big enough on X to experience anti-indian hate

we gotta party @cjrohandj @multimodali @navvye @VedantRath90232

[cj @cjrohandj]

@_anishlk Have u heard of the gadfly

[cj @cjrohandj]

@multimodali It’s nonlinear wouldn’t expect u to get it

[cj @cjrohandj]

This is an incredible display of mathematical talent and taste. I wouldn’t hesitate to say Anish is the LeBron of whatever this is. By the way, he’s single.

[anish lakkapragada @_anishlk]

We extend @cognition’s SWE-2 reward function to steer the Pareto frontier, not just improve it.

SWE-2 uses S − λₑC, where λₑ is cleverly the frontier’s derivative at that effort.

We show that adapting λₑ can target a desired score/cost improvement balance. See below!

[cj @cjrohandj]

Pray @oaklab gets RSI first. The Canadians will just use it to increase throughput at Tim Hortons and their healthcare system.

[anish lakkapragada @_anishlk]

We extend @cognition’s SWE-2 reward function to steer the Pareto frontier, not just improve it.

SWE-2 uses S − λₑC, where λₑ is cleverly the frontier’s derivative at that effort.

We show that adapting λₑ can target a desired score/cost improvement balance. See below!

[anish lakkapragada @_anishlk]

had so much fun with this blog @proximalhq! as models reach new frontiers, evaluations of cyber capabilities are critical for agent safety. through our audit of cybergym, we highlight fairness issues with vuln-repro tasks with verifiers such as differential execution

[anish lakkapragada @_anishlk]

announcing formalization progress on the BBFM conjectures!

i have formalized conjecture 6's unimodality claim for all n ≥ 2 and conjecture 4 for all n ≥ 2^(10^8). this work builds on @axiommathai’s prior work, closing problems left open 🧵

[cj @cjrohandj]

@_anishlk @axiommathai Investing in this tweet early

[cj @cjrohandj]

@zackkanter Bro you do B2B SAAS

[Skeletal Patriot @SkeletalPatriot]

@yuhline Here’s the full version posted by C-SPAN of the live broadcast not some conveniently edited version.

[Yuh-Line Niou @yuhline]

CNN got the extended video and showed how quick it was of a greeting.

[cj @cjrohandj]

@linamkhan Something else that’s worth mentioning is the way “nonprofits” in the space like AI eval companies are funded and interact with the AI model companies themselves. Model companies will try to say these evaluators count as “due diligence”, but they are funded by the same people.

[Lina Khan @linamkhan]

Law enforcers already have authority to charge companies and their CEOs for creating and releasing dangerous, unvetted, or defective products. We shouldn’t let discussions about new legal regimes distract from the fact that there’s no AI exemption from laws already on the books — a point @FTC emphasized repeatedly during my tenure.

1. There is an extensive set of laws that govern dangerous and defective products. For example, releasing unvetted AI models or agents can violate consumer protection laws. Shipping flawed AI tools without implementing adequate measures to detect and stop rogue or defective AI agents can be an “unfair or deceptive” act or practice under the FTC Act (and analogous state laws). And some state AGs are already exploring holding AI firms and their CEOs criminally liable when their models participate in criminal activity.

2. Existing laws also prohibit “unfair methods of competition.” This covers instances where AI firms appropriate the competitively sensitive information of their customers, including through tracking their use of various tools. It can also cover instances where firms pursue dangerous behavior, aware that doing so may compel rivals to do the same.

As the Supreme Court has noted: “A method of competition which casts upon one's competitors the burden of the loss of business unless they will descend to a practice which they are under a powerful moral compulsion not to adopt, even though it is not criminal, was thought to involve the kind of unfairness at which the [unfair methods of competition] statute was aimed."

3. The highly concentrated and interconnected structure of these markets could be creating major risks and conflicts of interest. We had started investigating these partnerships and cross-investments across the stack (and released a preliminarily overview of some findings: https://t.co/jJ5cS3Pin3).

Both federal and state enforcers should be scrutinizing these opaque relationships and inter-dependencies. We are already seeing how these relationships could undermine accountability. For example, OpenAI could face liability given the Hugging Face incident, but Hugging Face being bought up by Nvidia means that we’re unlikely to see it file a lawsuit over this — given Nvidia’s strong incentive to see OpenAI continue full speed ahead.

4. As AI tools dramatically change the landscape of cybersecurity risks and hacks, all businesses should be doubling down on having core security protections in place. Firms that fail to invest in adequate data security measures or fix known vulnerabilities can also be breaking the law. A recent analysis showed that around 1/3 of Fortune 100 companies do not even have a way to notify them about security issues. During my @FTC tenure, we sued firms for poor data security practices and held CEOs liable when they were personally responsible.

https://t.co/nwZ5Av8fOK

https://t.co/KjRye8y9SY

5. As policymakers consider new legal regimes, we should be looking to lessons from prior efforts to govern major sectors, such as banking and other networks, platforms, and utilities. Tools like structural separations, nondiscrimination, and supervision could be key, and there’s a rich history of what works and what doesn’t. But we can and must pursue any new efforts alongside enforcing existing laws.

[Adi @adi_baradwaj]

Timnit Gebru should be one of the independent evaluators

[cj @cjrohandj]

@sama The same guys who fund the model companies are funding the evals

[cj @cjrohandj]

@RichardSSutton What do you think about the situation? I’m sure many are interested in your perspective