amen

yesterday, had a conference #AISummitBarcelona I heard someone say, focus on EPOCH: empathy, presence, opinion (and judgement), creativity and hope. Loved the acronym coined by Isabella Loaiza, a recovered academic.

The loudest voices stoking fears about AI dangers have made tremendous headway in the past two weeks. AI technology has not taken some unexpected, dangerous turn, but the hype around it — propelled by what appears to be a well orchestrated PR campaign — has drummed up considerable fear. I worry that it represents a setback for our field. I have written frequently that fears of AI are overhyped. AI’s capabilities can be uncannily human-like and unpredictable, and it’s rational to worry when people who are directly involved express concerns. But I see the problems as a sign of the engineering work that ahead, rather than insurmountable barriers or the sky falling. AI technology continues to advance — which is a good thing! — but technical advances, poorly understood by the public, give those who seek to generate hype repeated opportunities to do so. First, I don’t see any step up in the risk of human extinction from AI compared to a few months ago. The theories about this remain the same fantastical, science fiction scenarios as a few months ago. The biggest change in AI risk is its cybersecurity capabilities — a topic which we should take seriously — but this, too, will not lead to the end of the world. The most notable recent event leading to increased fear was when an OpenAI team deployed an agent swarm that hacked into Hugging Face. Much of the popular press contained significant hype. For example, some publications reported that a swarm of 1,200 agents carried out the attack. While this was technically accurate, as I write this, I have about 1,300 processes running on my laptop. Yes, the ability to get large swarms of agents to work in parallel on a task is a significant technical advance, And, in computing, many processes run at the same time. So this shouldn’t be seen as some magical capability. Additionally, OpenAI’s buggy sandboxing and monitoring processes were key to enabling this incident. Fixing these bugs and putting in place improved monitoring would be appropriate fixes, not pausing AI. There are many well known ways to attack software systems. The main advantage of AI agents is that they are relentless. They will tirelessly try many tactics — and have the patience to chain vulnerabilities together — that previously would have taken an infeasible amount of human effort. But in the long term, I believe the advantage will lie with defenders (because they have more information with which to identify bugs, which they can fix), but the cyber-threat landscape has changed significantly. There are still bottlenecks to identifying and exploiting a vulnerability. AI agents still have to try a lot of things to see what works, and taking these actions takes time and might be detected by defenders. This is why, even though it is now easy to obtain versions of leading open weight models that have had their guardrails removed or weakened, so they will not refuse to try to execute cyber attacks, the world has not ended. I am also concerned about the anthropomorphization of AI in a lot of reporting, where LLMs and agents are unnecessarily treated as if they were people. If I wield a hammer, miss a nail, and accidentally dent the wall, it’s not the fault of the hammer. The problem lies in how I used the hammer. Similarly, if I prompt an agent and it hacks into someone else’s system, the responsibility lies with me, not the agent. Of course, we want to build systems that are as safe and predictable as possible. (For example, an unsafe hammer would be one whose head randomly flies off under normal use.) Today’s agentic systems are not predictable, but I see no reason why, by applying sound engineering practices, we won’t be able to make them extremely safe to use. One new element in the forecasts of AI-enabled doom is AI companies disclaiming responsibility for their own products. “I didn’t do it; my out-of-control agent did!” There’s a balance to be struck between the responsibility of the tool maker and the tool user, but when something goes wrong, let’s hold the people building and/or using the hammer responsible, rather than the hammer. (By the way, if you’re worried about AI bioweapon risk, David Bellamy has a great post on why this, too, is overhyped. Briefly, the bottleneck in building a bioweapon is not intelligence, but lab work and manufacturing.) Pausing AI progress will create much more harm than benefit. First, our adversaries will certainly not slow down. Second, engineering requires discovering problems empirically so we can fix them. If we pause AI by a decade, we will also delay finding and implementing safety engineering fixes by about the same duration. Of course, the incentive to stoke fears — for regulatory capture, to garner attention, or to make one’s technology seem more powerful — remains the same as before. Disclaiming responsibility is a new one. Taking a hard technical look at the actual risks however, I see little factual basis for the degree of fear that’s been stoked up. We still have hard research and engineering work ahead to improve AI safety, but the beneficial applications continue to vastly outweigh the risks, and we should keep building. [Original text (with links): deeplearning.ai/the-batch/is… ]

Imagine getting arrested for stealing Amazon packages in Miami and you show up to your trial and Jeff Bezos is a juror

Today, we're combining Cowork and Chat into one Claude. You can fire off questions and then turn to serious work in the same conversation. When your computer is open, Claude can use your local files and apps - and if you close it, Claude will keep working with its own computer in the cloud.

There are a few reasons why young people don’t like AI---including this.

yes, a friendlier UX for regular people. thank you

Hace un año, los Gobiernos de Francia y España propusimos establecer una edad mínima para acceder a las redes sociales. Muchos nos miraron con incredulidad. Otros se opusieron. Pero hubo Gobiernos que nos apoyaron y se fueron sumando a nuestra causa. Hoy, la Comisión Europea ha puesto sobre la mesa una regulación comunitaria con límites de acceso y seguridad por diseño. El mensaje que le mandamos a los tecno-oligarcas es claro: la infancia no se destruye con algoritmos ni intereses económicos. La infancia se protege con nuestras democracias.

this.

Hi @patrickc - Judith from @langdock here. We just crossed €50m ARR and flipped from a US Inc -> European SE. We want Germany to make starting and running companies easier (first on the list: compulsory notarization). At the same time, building a big company from here is absolutely possible and we intend to prove it. The SE is actually a good structure, but it's not an option for founders who are just starting out. That's why we are big supporters of @euinc. We're bullish on Germany - and bullish we can do even better. Both can be true at the same time. Happy to chat more.

couldn't agree more. Scientists, Founders, please all come back. We're relatively sane here and need you to stay relevant!

I genuinely hope that this is the start of a movement! Welcome (back!) @langdock

Has anyone resigned from a Chinese AI lab over concerns about AI safety/alignment yet?

The debate around recursive self-improvement is one of the most important, and contentious, in AI. I've heard people argue that the idea is nonsense: it’s mostly hype for impending IPOs. Lots of other people argue that yes, it's coming, but it’s dangerous, and we need to stop it. Then there’s a third group who believe that we’re quite close to RSI, and when we get there, it will usher in an era of discovery, productivity, and prosperity. @RichardSocher is in the last camp. He created You[.]com and then a company, Recursive Superintelligence, that is trying to make the tech a reality. He has a new book, “The Eureka Machine,” in which he describes his vision for swarms of agents working toward advances in everything from biology and medicine to chemistry and economics. This vision of RSI would be great for humanity, but I worry deeply about our capacity to get there. Smarter AI requires smarter oversight from within tech companies and from the government. The track record with both so far is not inspiring. So I sat down with Richard to talk through what a future with self-improving AI might look like. We talked about the risks—rogue systems, bioweapons, hallucinations—and what we can do to mitigate them through goal setting, regulation, and financial incentives. We touched on how intelligence can be broken down into units, why the limits of our own minds could hold AI back, and whether we’re projecting too many fears about human behavior onto an insensate technology. Richard is a brilliant man. I greatly enjoyed our conversation, and I hope you will, too. piped.video/watch?v=eqYqUAdj…

Perhaps it is worth stating that if the AI labs cannot organize what is essentially a fancier friend hang (a gathering of the top minds, representatives from all the labs + research institutions like METR + federal agencies, multiple days/weeks, a review of the recent cyber incidents, review of threats including cyberwarfare/biological warfare/electric grid interference, the beginnings of an action plan), then we should not trust them to do whatever comes after that. Get the best celebrity wedding planner in DC, pick a date, and figure this out. I can’t read another tweet from a well-meaning researcher saying “we have to get together and talk”. Ok then. Pick a date. 11-11 sounds good. It’s a Wednesday. And it’s the same for both the US and Europe, so there will be no confusion, even better. It emphasizes there is one shot to getting this right.

We've just raised 3B€ to scale our training and inference compute, and make open and sovereign AI the technology frontier. Grateful to everyone who got us there, excited by the fight ahead mistral.ai/news/mistral-make…

what a great read. Thank you @sapinker for sharing and @ProfCalNewport for writing this. Love both your work!

this is where non technical people also hit a wall.

Passive learning doesn't work for the same reason that watching an exercise video doesn't make you any fitter. You want to get fit? You have to actually perform exercises yourself. Same goes for learning.

such a great read. definitely worth of everyone's time.