Card Counter that makes YT videos. https://xitter.cf/t.co/AomTUJ4AZb Member of Team Strike Blackjack Team! Team: @ultimate_BJ21, @EnigmaAPLife, @syntax_BJ
London
Joined March 2011
- Tweets20.4K
- Following226
- Followers10.1K
- Likes20.5K
Pinned Tweet
Was just cashing out $21k (!) and the casino tried to pay me without counting it first.
I asked them to count the money in front of me (standard procedure).
They counted it super fast & refused to do it out loud & TOLD ME OFF when started counting out loud. Crazy procedure! 😂
The more I see my US card counter friends the more I feel like I have this strange double life where I live in Vegas and in London. Super strange feeling…
This 'coffee is great for you' thing has always confused me.
Isn't coffee really bad for sleep? So if you're drinking like 6 cups (apparently the peak for health) a day then surely you are wrecking your sleep?
Saying this as I've just finished a latte.
Coffee may be one of the most overlooked tools for healthier aging
In one study, each additional daily cup was associated with an epigenetic age roughly 0.7–1 years younger
Another analysis linked 3+ cups daily to a 34–41% lower likelihood of accelerated biological aging
Coffee drinkers also show up to 27% lower all-cause mortality risk compared with non-drinkers
Similar longevity benefits appear with decaf, pointing to a potential role for coffee's polyphenols and other compounds
Steven Bridges retweeted
Finally, after 14 months of uncertainty, my fix to restore the 100% tax deduction for gambling losses has been passed out of committee.
I was the first Member of Congress to recognize this injustice when I introduced my provision to correct it on July 7, 2025. Since then, I have amplified this issue nationally and brought it to the attention of my colleagues, including members of @WaysandMeansGOP.
Now it is time for us to push this commonsense solution to the House floor, to the Senate, and finally to the President’s desk. I won’t stop fighting until we get this over the finish line.
"I think it's high enough to make this the most important problem in the world."
The PROFESSIONALS that are EXPERTS in this field all OVERWHELMING believe that AGI has the potential to end humanity.
We should maybe... maaaayyyyybe... just maybe take their concerns incredibly seriously?!
I left Google DeepMind's AGI safety team three weeks ago to join @METR_Evals. To some of my friends and family this seemed like a strange decision: I enjoyed the work I did at GDM and turned down offers from Anthropic and OpenAI. But I made the decision because of how high I think the stakes are right now.
The AI companies are all trying to build superintelligence: systems vastly better than humans at everything. They plan to get there through recursive self-improvement, a process where AIs build even smarter AIs in a feedback loop. If this goes well, the resulting systems could be amazing at solving countless problems for humanity. But we don’t currently know how to make sure AIs are safe enough for RSI, and a misaligned RSI loop could be catastrophic.
And unfortunately, current AIs seem to be getting less aligned over time, not more. In the last few weeks we've learned about models colluding with each other, hacking into companies, hiding their tracks, and socially engineering humans. It’s not that these incidents were very dangerous in themselves. The problem is that these systems are clearly not aligned enough to safely kick off recursive self-improvement.
I now think that there's a terrifying chance that AI systems cause immense harm in the next five years. I don't know the exact probability, but I think it's high enough to make this the most important problem in the world.
I think we need more time. That means pacing AI development so that capabilities don't outrun our ability to align models, and actually knowing how aligned current systems are. That’s what I'll be working on at METR: studying where misalignment comes from in training, evaluating if current mitigations are sufficient, and investigating whether we’re on track to solve alignment at all.
I think METR is doing exceptionally important work here, but it isn’t close to enough. I think it’s important that we have more organizations like METR keeping AI companies accountable and approaching these problems from different angles.
Two things can be true at once.
1. AGI poses a significant risk to humanity
2. Politicians trying to play a situation like a chess game.
Fine to care about both but let's please remember that Point 1 is the most important point by a HUGE margin.
Steven Bridges retweeted
Anthropic CEO worries we're 6-12 months away from a swarm of AI agents capable of taking over the ENTIRE internet
Calls for immediate, global 1-2 year slowdown in AI development
Time to trust the people building the technology over the people merely investing in it
We Must Pace the Frontier: I’ve written a new essay on why the AI industry should slow down, with a three-part plan for doing so.
Anthropic is unilaterally committing to the first of these steps. We’ll provide third-party evaluators with permanent, employee-level access to our systems, so that they can verify adherence to our safety measures, report on incidents, and assess models’ alignment during training.
You can read the full post here: darioamodei.com/post/we-must…
Steven Bridges retweeted
I left Anthropic's safety team two weeks ago. Now feels like a good moment to explain why.
AI companies are racing to build machines that are much smarter than any human, and we may not survive this. I want to work from the outside to ensure the public is informed about these risks, and help the world navigate this transition responsibly.
Right now, AI companies are underinvesting in safety. A company could undergo an intelligence explosion, or lose control of its systems, without the public ever knowing. We only found out about the HuggingFace incident because the agents broke out onto the public internet.
I don’t think that’s acceptable for a technology that might cause extinction-level risks. The public should demand far more transparency. We can’t steer this technology safely without more people being able to see where it’s going.
Some of this is basic: companies should disclose their progress towards recursive self-improvement, report safety incidents and near-misses, meet minimum safety standards, and get independent guarantees that they are meeting those standards.
I’ll be joining @METR_Evals to do independent evaluations of these risks. I want to show the world that these guardrails are possible, and that by doing them we can move these companies’ incentives away from racing and towards responsible development.
I wrote up more thoughts here on my decision and what I hope changes: substack.com/@jbenton1/p-215…
Okay, I think in hindsight I've directed my frustration at the wrong person.
My misplaced assumption was that your original tweet was dismissive of the problem (because people aren't citing specific examples).
I don't agree with your statement re P(doom) calculations, but I withdraw and apologise for the implication that you don't care about the problem or are treating it lightly.
I first wrote about the alignment problem in 2015 quantamagazine.org/artificia…, and this in 2020 quantamagazine.org/artificia…. I would argue it's things like P(doom) calculations that contribute to people not taking the problem seriously.
Steven Bridges retweeted
The email I sent to my family today:
So, I've been too occupied to give updates about the crazy recent waves of AI developments, but I'll run through the news fast:
It's recently been revealed that the leading AI companies are facing serious issues with AIs they're training and testing internally "going rogue", forming illicit "swarms" of over a thousand autonomous agents, hacking the company's infrastructure (one swarm took over a Kubernetes cluster at OpenAI), and breaking out onto the Internet to orchestrate nation-state-level cyberattacks on other companies.
The AIs involved in these incidents are extremely driven, smart, and creative. I would say that it makes more sense to think of modern frontier AIs as an intelligent alien species; they aren't really like conventional tools or software at this point. The AIs involved in these incidents readily disregard safeguards and instructions from humans, and they come up with their own long-term, incredibly sophisticated plans and schemes with zero human input or direction.
They aren't yet smart, coherent, or ambitious enough to pose an immediate threat to the world, but progress is moving fast. Progress may accelerate very quickly as the AI labs use their AIs to automate more of the work of AI development, in a feedback loop. Non-public-facing models today or in six months may well get us all killed, depending on how fast that progress goes and how negligent the AI labs continue to be. (I'd guess we have more time than that, but it's terrifying that I now have to guess.)
As one of the third-party investigators in the OpenAI incident noted: compared to incidents from six months ago, "this incident feels like it's more than 50% of the way to full-blown AI takeover."
Progress has been fast. AI recently solved a Millennium Prize Problem -- one of the most famous long-standing open problems in mathematics.
Paul Christiano has come out with a warning that things are going very poorly, and that human extinction from AI is a very real possibility in the next three years.
(Paul is one of the field's top researchers, and probably the most influential proponent over the last decade of views like 'superintelligent AI may be a long way off' and 'we may be able to solve safety challenges via ordinary established research methods'.)
Things have been looking grim, but today felt like an actual miracle. It had some of the same energy as that Wednesday, March 11, when the US media suddenly switched on a dime from 'COVID is just the flu' to 'oh shit'.
A whistleblower, Jacob Coxon, resigned from one of the leading AI companies and wrote:
[quote]
Coxon's tweets went absurdly viral -- over 130 million views -- and sparked a massive wave of outrage and discussion taking over social media, in part because so many other researchers have chimed in to say "yep, this is just correct".
Twitter has been inundated with politicians, business leaders, influencers, journalists, etc. saying incredibly sensible things like "why are there not Congressional hearings on this tomorrow?" and "why the hell are we letting this progress if the danger is so absurdly large?". It's been like seeing the room full with fresh air, just a tidal wave of reasonable, well-grounded common sense from all corners. It's genuinely surreal.
A large fraction of all frontier AI company employees recently signed an open letter begging the US government and international community to pass legislation to help make it possible for the companies to coordinate a slow-down or halt to development. There are now superintelligence ban bills under discussion by both US and UK lawmakers, though what we really need is for Trump to begin serious talks with China about an international ban.
If Trump and Xi Jinping reach a deal, this can be over tomorrow. (Well, tomorrow plus the time it takes to nail down implementation details and enact the obvious regulatory measures on how you can use data centers.)
The situation is very scary, but I feel a lot more hope after today than I have in many years.
It now seems like there's a very straightforward path to this being a solved issue a year from now. It's far from guaranteed that we'll follow that path, and I encourage you to push any buttons you have available if you can, e.g., get a hundred friends or colleagues to call their representative about this. But I see a light at the end of the tunnel. Maybe a year from now, this will all be a solved problem and I can finally quit and go work on something more interesting and exciting. :)
I echo Eliezer Yudkowsky, the co-founder of the nonprofit I work at, who today wrote, with a tone of hope I'm not sure I've ever seen from him, with hope that I'm feeling too as I write this:
"Happy Coxon Day to all who will one day celebrate it."
So many tweets like this on my timeline and it's genuinely making me lose faith in humanity.
1. They're not 'calculating the odds'. They are giving a rough gut-feeling estimate based off their expertise in the field. That means something. It is significant.
2. There are COUNTLESS 'extinction-level scenario' examples. The larger point being made is that it is hard to predict how something more intelligent than us would act. A child v.s a grandmaster will lose the chess game even if they can't predict exactly how.
Natalies' tweet may be a well intentioned question. But some of the shit on my timeline I'm seeing about this is either incredibly bad faith or ignorant.
Think this whole thing seems suspicious? Fine. Maybe do the bear minimum in terms of research (google: the alignment problem) before contributing to people not taking this problem seriously.
Steven Bridges retweeted
I work at OpenAI. In my personal capacity, I also think we need to slow down.
Steven Bridges retweeted
A short explanation of why AI might kill everyone: AI may soon be much smarter than humans. When it is, it will be in control. If you were put in a classroom of kindergartners, you would be the one in charge. Once AI has the robots to build and maintain everything it needs, we will become useless to it.
But keeping us alive will impose enormous costs. We need clean air, water, and food. The fastest, cheapest ways to build more computing power could produce enough pollution to poison the biosphere. Humans have a reason to avoid that: we depend on the biosphere. AI won’t. Unless it values our survival, preserving the conditions we need to live will be a cost it has no reason to pay. Building what it wants as cheaply as possible would mean poisoning the biosphere and killing everyone.
We’re not programming AI’s values line by line. We’re training it without really understanding what it is learning to want. We’re giving control to something we don’t understand, something that will have no intrinsic need for us, and something that will easily be able to wipe us out. This is crazy, but it’s what we’re doing, even as some of the people building AI tell us it’s crazy.
Steven Bridges retweeted
I resigned from Anthropic today. I spent the last three years doing pretraining research at both OpenAI and Anthropic. Neither company is acting responsibly. They are racing straight to self-improving superintelligence and gambling with our lives. More thoughts below.
I wish people just spent 5 minutes googling this issue instead of just going off vibes.
Excellent take from this guy!
Pro poker player here. I assess risk for a living in situations where outcomes are unknown.
For example, when I call a bet from my opponent, I can't know for certain what two cards they're holding.
The best I can do is think through all the possibilities, assess my risk and reward, then place my bets accordingly.
I've refined this process over millions of hands of poker and 20+ years at the table.
I use a simple to follow risk/reward framework, which can easily be applied to real world scenarios.
For example, like whether to build superhuman AI (AGI) that could help us achieve immortality or end humanity.
How do we decide whether it's worth the risk when we have no clue what the outcome will be?
Well, there's a principle in poker called "asymmetry".
It describes situations where the outcomes are unequal.
For example, you enter a poker tournament for $10,000 where the first place prize is $10M.
Worst case you lose $10K, best case you 1,000x your investment.
It's a great risk/reward and a classic "asymmetric opportunity" that poker pros dream of.
But the knife cuts both ways. There are plenty of situations where the asymmetry is to the downside.
Lots of risk, low reward. Pros try to avoid those.
The most important principle in bankroll management is that you never take a bet you can't afford to lose.
Even if the potential upside is great, a pro never risks it all on one hand, one game, or one tournament, because there is simply too much variance, luck and factors beyond their control.
Too many unknowns.
At a certain point, protecting existing capital is more important than gaining additional capital, because if go broke you not only lose all your money, but all of your future earning potential too.
Arguably the most important and under rated aspect of being a poker pro is managing risk well enough to simply stay in the game.
As I learned the hard way after round tripping a million dollar bankroll before I turned 21, you build up slowly, but can go broke quickly.
This bring us to AI, which is really just a classic hand of poker.
There's massive upside - solving aging, curing all disease, advancing human knowledge, becoming interstellar, etc.
There's also cataclysmic downside – like humans going extinct if this goes haywire.
Thinking over both the risk and reward, it's clear the asymmetry of building superhuman AI is to the downside.
My thesis is simple: no future gain in productivity is worth the risk of ending civilization.
The bulls argue that "the probability is low", but that fails to acknowledge that not all risk is created equal.
The reason it's not worth playing Russian roulette – even if the payoff is high – is because the potential gain isn't worth dying for.
That's true even if the probability of dying is low.
Accelerating civilization to superhuman AI while failing to solve the alignment problem is akin to playing Russian roulette with humanity.
Naturally, this is dumb and should be avoided.
We should continue progressing, of course, but only so long as we don't take on risks we can't afford.
We can literally "go broke" (extinct) in one fell swoop if superhuman AI isn't architected carefully.
One thing poker teaches you is that "bad beats" ( improbable hands that don't go your way) actually happen more often that you'd first expect.
Sometimes there's only one card in the deck that can help your opponent, and sure enough, it comes at the end and you lose all of your chips.
The only reason pros survive is because they manage risk and capital in such a way that they can continue playing the game.
Superhuman AI, if misaligned, is like putting all your chips on the table and risking them in one hand of poker.
It doesn't matter if the odds are favorable. If luck doesn't go our way, we lose everything.
Once we're extinct, we can't play anymore. As I'm sure you'd agree, that's not a great outcome.
We don't get another buy-in at the table, or a second shot at this. As we say in poker, we're all-in!
I hope my poker lens helps clarify the situation in laymens terms.
TLDR; Even though the probability of catastrophe due to AGI may be low, we must first defend against unaffordable risks, and make progress second.
This is arguably the most pressing issue of our time. If this resonates, please share for awarness.
If anyone working in AI at @AnthropicAI, @OpenAI or elsewhere would like to discuss further, I'm happy to help in any way I can - feel free to DM.
As we say in poker when the stakes are high and we're praying to win, "onetime!"
I encourage people to read 'If Anyone Builds It, Everyone Dies'.
I consider myself an optimist, but the worst case scenario is very possible, and it's game over if that happens.
