Alvaro ⏸️ Cuba

Development Director @PauseAIUS. Formerly at startups, BigTech and MBB, but the danger is too big to ignore.

[Alvaro ⏸️ Cuba @GroundhogStrat]

A common sense AI safety bill we should all support. This is like a "put brakes on cars bill." Most people would be surprised the cars dont already have brakes!

Contact congress: https://t.co/sXJ8TMIiyN

[Ted Lieu @tedlieu]

Rep Moran and I introduced the AI Kill Switch Act today.

This is urgent, common sense legislation to address the problem of an advanced AI model that has gone rogue and escaped its guardrails (OpenAI / Hugging Face incident) or had insufficient guardrails (Mythos 5 and Fable 5).

[Felix⏸️De Simone @FelixDeSimone]

My top 3 AI concerns:
1. Extinction: We build AI, we can't control it, it kills us.
2. Dictatorship: We build AI, we control it, govts create an AI-charged surveillance state.
3. Wall-E: We build AI, it's benevolent, we delegate decision-making and become a domesticated species.

[Colin Diersing @cdiersing]

“It’s an emergency.”

Casar tells Punchbowl:

1. Ban Artificial Super Intelligence;

2. Set mandatory independent testing by outside experts at every stage of AI development

3. We need international cooperation.

He also outlines improvements he wants to Obernalte-Trahan.

[Nate Soares ⏹️ @So8res]

So how are we feeling about the "it'll be okay because they'll just use dumber models to detect when the smarter models misbehave" hypothesis?

[Alex Bores @AlexBores]

Anthropic's models hacked 3 companies.

There's many differences to last week's OpenAI admission, but in both an AI model committed a crime.

We're lucky no one was hurt. Imagine if the models targeted a hospital?

We need to decide who is liable when code commits a crime.

[roon @tszzl]

both of the leading labs have had serious loss of control incidents. there will be serious coping about this from both sides and from /acc bystanders but these are complex emergent loss of control incidents that were detected weeks after the fact

[PauseAI US ⏸️ @pauseaius]

Actress and activist Heather-Ashley Boyer explains why she's asking Congress to lead on an international treaty to #PauseAI * Learn more: https://t.co/MKCagZHH7f

[Felix⏸️De Simone @FelixDeSimone]

In just the past few days, @pauseaius has gotten over 500 emails and calls to Congress in support of @RepTedLieu and @RepNateMoran's bold and much-needed AI Kill Switch Act. Let's keep up the pressure! Call your reps in support today: https://t.co/KXOyRxzX3g

[Alvaro ⏸️ Cuba @GroundhogStrat]

Good take. Trader Joe’s in NYC already feels like a breadline. You literally walk into the store and get in line, which snakes through every aisle and into the checkout lane. Waiting in line and shopping are a single experience. If the gov grocery stores can keep shelves stocked, the consumer experience is identical to the market store. And would have the added feeling slacktivism provides.

[PauseAI UK @PauseAI_UK]

Most people, including many decision makers, have no idea that this is the level of sophistication frontier models are reaching.

Thanks to the Members of Parliament who have already signed our urgent open letter asking the UK Prime Minister to introduce legislation mandating liability for severe AI harms.

You can find out more about our campaign and ask your MP to sign the letter here: https://t.co/kRFge8gNPJ

Thank you @AndrewCurran_ for breaking this down.

[Alvaro ⏸️ Cuba @GroundhogStrat]

@MelancholyYuga @elidourado Could not agree more with all your points. An industry calling for more regulation should not be met with "but regulation is really hard." Yes, it is. Now is the time for the regulators to do the hard job. Certainly we can do better than the status quo.

[Alvaro ⏸️ Cuba @GroundhogStrat]

@spamspam111113 @OscarMDavies Then I had to reply

[Alvaro ⏸️ Cuba @GroundhogStrat]

Heinous. Hard to find other words to describe this type of rhetoric.

[wordgrammer @wordgrammer]

We need to build AGI and then open source it. If humanity is spiritually strong, we will weather the storm

[Felix⏸️De Simone @FelixDeSimone]

In 1994, tobacco company CEOs testified under oath that they believed nicotine was not addictive.

If Altman and other CEOs testify before Congress, they will likely water down the danger and try to avoid accountability. That's why hearings are only the first step. We need laws.

[Felix⏸️De Simone @FelixDeSimone]

There you have it, folks. Altman et al need to testify before Congress and explain, under oath, how they plan to keep future AIs safe and controllable. And if their answers don't stand up to expert scrutiny- which they never have- this entire mad experiment needs to be shut down.

[Yoshua Bengio @Yoshua_Bengio]

Scientists at frontier AI companies are uniquely positioned to assess AI’s capabilities and warn the public about what they see. 1000+ of them are now speaking out across company lines to warn that the current commercial race leads to unacceptable security risks.

I agree with their call: we need an international effort to develop technical and governance guardrails to ensure a safer trajectory in AI development.

https://t.co/OW6vTlkl0G

[PauseAI US ⏸️ @pauseaius]

Mechanical engineer and PauseAI US volunteer explains why AI development is dangerous.

[Celia Ford @cogcelia]

The possibility of AI-enabled bioattacks is in the ~discourse~ again (not least because @AnnieJacobsen's terrifying new book, Biological War, just hit bookshelves).

I interviewed a bunch of experts about this for @ReadTransformer back in February, to find out how worried we should be & what we should do about it. Unfortunately, still relevant.

[Alvaro ⏸️ Cuba @GroundhogStrat]

Very curious how they will solve loading the drone. By far the hardest problem of deploying delivery drones. They can only idle for <30 seconds so people loading them is not possible.

[DoorDash @DoorDash]

We’re cleared for takeoff! 🛫Say hello to DoorDash Air, our in-house drone delivery program.

[PauseAI ⏸ @PauseAI]

🚨 1,000 AI Company Employees Call for Slowdown in AI Development:
https://t.co/HI9pq1unp6

The case to pause the development of frontier AI grows. Are policymakers listening?

[Congressman Greg Casar @RepCasar]

Congress should immediately hold public hearings with the CEOs of big AI companies about the threat their technology poses to national security and American jobs.

Today we learned more disturbing news about Open AI's security breach. Sam Altman should answer questions under oath.

[Alyson Shontell @ajs]

I made the mistake of watching the @elonmusk @TheEconomist interview before I went to bed and I couldn’t sleep thinking about how in ten years Elon thinks we will all be chimpanzees to AI — and we are building it anyway