Brad Carson

Father of Jack, husband of Julie. Ex Congress, DoD, @BattenUVa, Prez @utulsa. Prez of @americans4ri. An enthusiast, but w/ a gimlet eye on the log x-axis. Okie!

[Brad Carson @bradrcarson]

"In a mindless age, every insight takes on the character of a lethal weapon." - M. McLuhan to E. Pound, letter, 1951.

[WIRED @WIRED]

Leading AI labs, executives, and scientists are sending a letter to lawmakers urging them to improve tracking of synthetic DNA sequences that could be used for bioweapons. https://t.co/vjOYASEY9M

[Congresswoman Valerie Foushee @ValerieFoushee]

In light of recent reports surrounding AI policy frameworks that preempt states from further regulating AI, I want to be very clear that I oppose any and all efforts to do so.

As a leader in the House working towards AI policy that protects working Americans, I understand the importance of allowing local leaders to further regulate this technology.

Democrats must be leaders in passing AI policies that put people first. Preempting states from passing their own AI regulations does not do that.

[Nathan Calvin @_NathanCalvin]

OpenAI and Sam Altman have no plans to make political donations this cycle - unlike OpenAI’s president (advised by OpenAI’s head of global affairs). But that has nothing to do with OpenAI. They are actually the victim here! Appreciate the clarification from Sam.

[Brad Carson @bradrcarson]

For people (like me) trying to really understand how the Palantir/Maven/Claude integration works in nat sec, highly recommend this @CSIS new report. The best explanation I've seen, alongside the OS reports from Israel about Gospel/Lavender (not Palantir, but probably some similar skills).
https://t.co/naNnjHK5ay

[Atlas Press @realAtlasPress]

Leonardo da Vinci's greatest paragraph

[Brad Carson @bradrcarson]

Wow. The chutzpah of this! They hired the self-proclaimed "master of dark arts" and openly tried to take over US AI politics (alongside a16z). That the project has been a PR disaster for them doesn't allow them to claim victimhood.

[Emily Birnbaum @birnbaum_e]

.@sama says that he would like to see money out of politics.

He said he does NOT plan to make any political donations this cycle.

But he added: If OpenAI’s competitors are “trying to use money to gang up on us, we have to be able to fight back.”

[Brad Carson @bradrcarson]

I had missed that they'd used @alltheyud picture. Even crazier. It's like the Keystone Cops COINTELPRO!

[Simon Lermen @SimonLermenAI]

Especially damning: pretending to be a "doomer" and using a profile picture similar to @allTheYud and then replying to John Sherman calling for violence.

[Michael Cohen @Michael05156007]

GREG BROCKMAN, MARC ANDREESSEN, AND BEN HOROWITZ SPONSORED A FALSE-FLAG TWEET CALLING FOR VIOLENCE AGAINST THE AI INDUSTRY

https://t.co/q49HE7RV4x

[dylan matthews 🔸 @dylanmatt]

One of the funniest sentences I’ve read in a while

[Brad Carson @bradrcarson]

@krishnanrohit Haven't read it yet. But would be grateful to understand how his LLM understanding is dated. Curious.

[Alec Stapp @AlecStapp]

CAISI's current operating budget is only $15 million.

CAISI needs at least $84 million annually to fulfill all AI Action Plan taskings related to AI readiness.

In other words, for the cost of a single F-35A joint strike fighter jet, the US government can gain situational awareness on the most strategically important technology of this century.

[Americans for Responsible Innovation @americans4ri]

🚨Ahead of tomorrow's NDAA markup, natsec and AI leaders are calling on lawmakers to include restrictions on AI-enabled autonomous weapons.

The letter follows calls from VP JD Vance to keep decisions on use of lethal force in human hands in the AI era.
https://t.co/Z6E7Jx9dUs

[Former Congresswoman Marjorie Taylor Greene🇺🇸 @FmrRepMTG]

Obernolte was a big supporter for the 10 year moratorium on state rights to regulate and make laws on AI and tech.

Big tech donors want it.

Will need to watch this one.

[Brad Carson @bradrcarson]

Crazy that “The Beauty and Joy of Programming has 1/3 failure!

[Litong Deng @_litongdeng]

Students are failing UC Berkeley CS classes at an alarming rate. More than 35% of students failed CS 10, a course described as “a gentle but thorough introduction to computer science.” In the past few semesters, less than 10% of students failed the class.

[Brad Carson @bradrcarson]

Actually watch all these games and didn’t know Mia was White Chocolates daughter. He was a great one! Nice to have him in Oklahoma.

[Ohio Divided @BuckeyeNatty]

https://t.co/jQIg6qtV89

[Americans for Responsible Innovation @americans4ri]

This is smart policy. ⬇️

The Sectoral AI Governance Act gives agencies clear authority to enforce the law when algorithms contribute to real-world harms. As AI impacts housing, health care, and basic rights, this is a good step forward.

[Americans for Responsible Innovation @americans4ri]

This week the President signed an EO enabling government oversight of advanced AI. It's the strongest action yet from the Admin to address AI risks.

Today on the Hill, ARI joined @secureainow and a bipartisan coalition calling on Congress to make those safeguards mandatory.

[Alex Bores @AlexBores]

I'm honored to have @CUFFHAction’s endorsement.

This organization has spent more than a decade doing the hard work of keeping families in their homes and in their communities. That's exactly the kind of fight I'm taking to Congress. The CUFFH Action community knows firsthand what it means when Washington gets housing policy wrong. I'll work every day to make sure the voices of tenants, immigrants, and working families in New York are heard at the federal level.

[Brad Carson @bradrcarson]

Just read through @oai new federal framework for AI regulation. Here's a hot take: It is what I expected, having seen their lobbying at federal and state levels. It has lots of good things in it re CAISI funding, whistleblower protection, and more.

It is also, IMO, not quite acceptable as a federal framework.

OpenAI's framework mostly federalizes what they're already required to do under Illinois law. No new burdens. Just a federal ceiling where state floors used to be.

CAISI in the OAI plan can evaluate models but never block deployment. Even the Trump administration has recognized that truly dangerous models shouldn't be releasable without scrutiny. If we are limiting federal review to the most dangerous capabilities - cyber, CBRNE - it only makes sense that, if a model can provide meaningful uplift, its release be blocked until sufficient containment.

And preemption remains underspecified. Most proposals for preemption block every AI-specific state law regarding labor rights, creator rights, child safety rights, sexual assault survivor rights, and many more. Laws that "relate to frontier development" could be read narrowly or broadly. The former is essential.

Finally, OAI does not require independent standards for audits and safety plans. To their credit, they do mention that such standards should be developed, but not sure to what end or timeline. This is the critical issue. A world in which a company can write the safety plan, grade it, and then ask an auditor to verify compliance is not a good one. The safety plans must be reasonable (language struck from the RAISE Act, alas) and should be based on required items, and the audits should not be mere compliance projects but substantive reviews. OAI hints at this at some future date, but this is a little unclear to me.

All in, nice to see OAI offer this. Lots of good things, but not quite right on the important issue of independent evals.

[Allie Maloney @AllieMaloney1]

Privacy advocates must hold the line on FISA to prevent a reauthorization without closing the data broker and S702’s backdoor search loophole. Pulte could be a leverage point. AI’s enablement of mass surveillance is too big of a risk to ignore.