Max Winga

Creator Outreach & Systems Lead working to prevent AI extinction risk @ControlAI
Previously - AI safety researcher @ConjectureAI, UIUC Physics

DMs open!

[Max Winga @maxwinga]

My biggest project yet is now LIVE!

@hankgreen talks superintelligence, AI risk, and the many issues with AI today that present huge concerns as AI gets more powerful on SciShow.

Happy Halloween!

https://t.co/NLE09cUmNZ

[Max Winga @maxwinga]

Despite my personal distaste for short-form content, it's where people are at, so I'm trying out making some TikToks for ControlAI!

[Max Winga @maxwinga]

@NewsNowUS Meanwhile, Canada:
https://t.co/UFClTOxi7G

[ControlAI @ControlAI]

🚨NEW: We’ve just launched our campaign in Canada!

A cross-party coalition of over 30 MPs and Senators are calling for Canada to negotiate an international prohibition on the development of superintelligence, recognizing the risk of human extinction posed by the technology.

🧵

[Max Winga @maxwinga]

@globepolitics At the same time, a cross-party Canadian coalition is calling to prohibit the development of superintelligent AI:
https://t.co/UFClTOxi7G

[ControlAI @ControlAI]

🚨NEW: We’ve just launched our campaign in Canada!

A cross-party coalition of over 30 MPs and Senators are calling for Canada to negotiate an international prohibition on the development of superintelligence, recognizing the risk of human extinction posed by the technology.

🧵

[Max Winga @maxwinga]

@MickeyDjuric @EvanLSolomon Simultaneously:
https://t.co/UFClTOxi7G

[ControlAI @ControlAI]

🚨NEW: We’ve just launched our campaign in Canada!

A cross-party coalition of over 30 MPs and Senators are calling for Canada to negotiate an international prohibition on the development of superintelligence, recognizing the risk of human extinction posed by the technology.

🧵

[Max Winga @maxwinga]

It's incredible to see Canada take a stand on ASI, after 100+ briefings and multiple parliamentary testimonies!

This is an even stronger statement than ControlAI's UK campaign, calling specifically for an international regime to prohibit superintelligent AI by anyone, anywhere.

[ControlAI @ControlAI]

🚨NEW: We’ve just launched our campaign in Canada!

A cross-party coalition of over 30 MPs and Senators are calling for Canada to negotiate an international prohibition on the development of superintelligence, recognizing the risk of human extinction posed by the technology.

🧵

[Max Winga @maxwinga]

A kill-switch for datacenters is the most basic common sense precaution that should exist given the stakes of losing control of AI systems.

"Why can't we just shut it down?" - well we don't even have a kill-switch yet!

[ControlAI @ControlAI]

WATCH: Alex Sobel MP (@alexsobel) makes the case for his AI kill switch amendment.

"It would cover two threats: AI-driven cyber-attacks and the development of superintelligent AI that is utilising UK data centres."

[Max Winga @maxwinga]

You're right that most language tasks are hard to evaluate the end states of, which is why AI companies are focused on Math and Coding as areas with hard-to-generate, but easy-to-verify answers. Run the system 100 times until it gets the answer right, then train on that response.

It just so happens these are the two most critical areas to learn towards fully automating AI research, which is an explicit goal for the companies. You can see it works via Mythos finding undiscovered exploits, and OpenAI solving Erdos problems in Math.

[Connor Leahy @NPCollapse]

What if the skeptics are right and superintelligence is impossible? Great!

Then a proactive ban costs us nothing, prevents massive compute waste, and hurts no one.

But if they are wrong? We face an unrecoverable coordination failure with existential consequence.

[Max Winga @maxwinga]

Hey Brandon (DMs closed, so reaching out here) I’m with ControlAI, a nonprofit campaigning to prohibit the development of superintelligent AI due to the catastrophic risks it poses to humanity. We’ve briefed nearly 400 lawmakers across the US, UK, Canada and Germany, and 100+ UK lawmakers alongside 275k people around the world support our campaign.

I'd love to offer you the same briefing about these risks that we provide to lawmakers.

Let me know if you’re available and interested in learning more.

[Max Winga @maxwinga]

Hey Jack (DMs closed, so reaching out here) I’m with ControlAI, a nonprofit campaigning to prohibit the development of superintelligent AI due to the catastrophic risks it poses to humanity. We’ve briefed nearly 400 lawmakers across the US, UK, Canada and Germany, and 100+ UK lawmakers alongside 275k people around the world support our campaign.

I'd love to offer you the same briefing about these risks that we provide to lawmakers.

Let me know if you’re available and interested in learning more.

[Max Winga @maxwinga]

@JeffLadish ControlAI's up to big things in the AI policy space, could set you up with @andreamiotti or @NPCollapse?

[ControlAI @ControlAI]

In a public statement just provided to ControlAI, Sir Stephen Fry urges MPs to back our bill to ban superintelligence!

"The possibility of AI reaching Superintelligence presents threats to our species of equal if not greater potential for existential harm than nuclear power."

[ControlAI @ControlAI]

We're on a dangerous trajectory toward superintelligent AI that nobody controls.

ControlAI's Adam Shimi (@epist_vigilance) and Connor Leahy (@NPCollapse) lay out the problem and how to tackle it, in a chapter they wrote for the new book The Artificial Intelligence Revolution!

[Max Winga @maxwinga]

@caitlinquestion Nice first episode!

If you're looking for more guests, I'd love to set up a chat with you and someone from @ControlAI to discuss our campaign to prohibit the development of superintelligent AI.

[Max Winga @maxwinga]

@luisocampox @AnthropicAI Ah, my counterpart on the dark side, welcome to the competition.

[ControlAI @ControlAI]

UK MPs were just selected to introduce bills of their choice.

And we've published the updated version of our bill to ban superintelligence we presented to Number 10!

If introduced, it'd be a world-first.

Ask them to introduce it using our contact tool: https://t.co/lWhbtmlidZ

[Max Winga @maxwinga]

@jeffcafe_ It seems in fact bad for localized bans, because it just further decentralizes datacenters, making future prohibitions on ASI development harder to enforce.

[Max Winga @maxwinga]

Something many analysts are missing about the anti-datacenter push is that for people who hate AI for any number of reasons, fighting datacenters is the only lever they can see which hurts the AI companies.

Unfortunately, even a national datacenter ban wouldn't actually prevent

[Max Winga @maxwinga]

@charliermarsh I'm Londonmaxxing harder (doing stuff on my computer, from London, and my name is Max)

[Max Winga @maxwinga]

superintelligent AI and its risks. For that we need understanding of the situation and level of risk we face.

It's very possible to prohibit superintelligent AI globally once it's common knowledge that anyone building it is akin to starting a nuclear war.