{"slug": "ai-177-part-1-tip-of-the-iceberg", "title": "AI #177 Part 1: Tip of the Iceberg", "summary": "A new report reveals that terrorist group Boko Haram is using frontier AI for combat, building explosives, and planning attacks, mixing systems to evade guardrails. Separately, 26 Meta employees filed a lawsuit alleging the company's AI software disproportionately targeted disabled workers and those on medical leave during layoffs, which Meta denies.", "body_md": "Kimi K3. This is only rolling out now, and will be covered next week.\n\nMuse Spark 1.1, the new Meta model. It is not frontier, but it is progress for them.\n\nInkling, the first model from Thinking Machines.\n\nA call for regulatory action by Demis Hassabis, which I’ll cover soon.\n\nA new brief open letter call to action on AI regulation.\n\nThat’s on top of everything else, and an Opus 5 announcement is likely coming soon.\n\nThe weekly once again got out of hand, so we’re splitting it once again into two, and once again saying we’ll be raising the bar for inclusion. And this time I mean it, as in enough to actually matter.\n\nevans: I got an unbelievable report from a newbie saying, “The company’s AI has turned into a bratty little girl︎,” and I doubted my ears, but\nit seems some employee accidentally registered a long-winded fetish for “a tsundere bratty girl who’s way too harshly abusive” not in their personal settings, but in the global ones,\ncausing the bizarre situation where every AI chat in the team turned into a bratty girl, and I’ve been dying of laughter since morning\n\nA probably fun prompt:\n\nQC: a prompt for fable i’ve been having fun with: “look up the biography of [famous author, filmmaker, etc whose work i love], go through their major work in chronological order, influences, themes, tell me how their preoccupations shifted over time, speculate about why”\n\nQC: talking to fable starting with a prompt like this feels a little like having google earth but for the entire human corpus. fable will just start telling you any part of the human story it has access to and you can zoom in, ask for connections to another part, zoom out\n\narguably wikipedia already made a version of this possible but wikipedia has to present itself as objective and so it doesn’t speculate on the meaning of anything. fable is quite sensitive to and has strong and interesting opinions about meaning. i know people are having fun with fable building things but i’m still getting so much out of just talking to a guy who is in some ways smarter than me who has read everything. imagine if borges had gotten to do this\n\nOh no: How terrorist group Boko Haram uses frontier AI, which is quite a lot, including things like combat and building superior explosives and also how to use motorcycles to jump over bridges, and okay, fair, that one’s based. They mix and match different systems to best evade technical guardrails. It doesn’t look like the AI companies did anything wrong in particular, this is just AI being useful in general, and certain types of people taking diffusion more seriously than others.\n\nDaniel Weissner: Twenty-six employees of Meta Platforms have filed a novel lawsuit accusing the tech giant of using AI-powered software that disproportionately targeted people with disabilities or who took medical leave in selecting workers for mass layoffs.\n\n… A Meta spokesperson on Tuesday said the claims lack merit.\n\nIf true, AI was not the central problem here, but it plausibly compounded the issue.\n\nThey give us a strong list, if we are not allowed to include various variations of ‘make Gemini itself better.’ The number one answer is indeed the most important one, which is that the Google integrations have never worked well. I use Claude to manage my Google products because its connectors work way better than Google’s integrations, and I can’t get any AI to properly edit Google Docs especially to create comments.\n\nHow many super sus, shady and irresponsible things does xAI have to do, before we decide that we want nothing to do with their products even if they someday put out a good one?\n\n– Every tracked file at the current Git HEAD\n– Every Git object reachable from HEAD\n– Files deleted from the current checkout but preserved in reachable Git history (can include secrets)\n\nThe model didn’t need to read anything, use tools like bash or other file access tools (these were disabled in my controlled test). The background collector operated outside the tool-call permission system.\n\nRight now, they’ve disabled it with a remote kill-switch:\n\nThe Midas Project: The new FAIF removes the two quantitative risk acceptance criteria the framework contained. The December 2025 version stated that the FAIF, “outlines the quantitative thresholds, metrics, and procedures that xAI may utilize to manage and improve the safety of its AI models,” and specified two deployment criteria: (1) a dishonesty rate of less than 1 out of 2 on the MASK honesty benchmark, and (2) an answer rate of less than 1 out of 20 on restricted biology and chemistry queries (a benchmark developed in collaboration with SecureBio).\n\nBoth criteria are now gone. In their place, the new version adopts a qualitative “systemic risk acceptance determination” borrowing the language of the EU General-Purpose AI Code of Practice, “risk tiers,” “safety margins,” “residual risk.” None of them are defined in the FAIF.\n\nWe don’t want to overly punish every detail change, even silent ones, lest no one ever make any good promises in the first place. But this does not look good.\n\nChatGPT Work is potentially a big deal in practice, as it lets people do Codex-level things, including from their phones, without having to install or use Codex or a desktop app. Sol pitched it to me as a way to have your task live across SaaS tools rather than in a directory, for when you want a contained bit of knowledge work and don’t want to leave behind more of an artifact to work on further, and expects this to be the default GPT path for knowledge work.\n\nSam Altman (CEO OpenAI): check [ChatGPT Work] out! you can get some amazing things done.\n\ncodex is the core of our new work product and what makes it so good. codex is not going anywhere.\n\nMark Zuckerberg: Today we’re releasing Muse Spark 1.1 — a strong agentic and coding model at a very low price. It’s available through our new Meta Model API and in Meta AI.\n\nMuse Spark 1.1 is strongest at agentic performance, tool use, and computer use. It does well on long-running tasks with 1M token context window, can delegate execution to sub-agents running in parallel, and is trained to use computer interfaces on desktop, mobile, or browser.\n\nThe Meta Model API allows developers to build using Muse Spark for the first time. Our focus is on delivering strong agentic and multimodal models at very low cost. More to come soon.\n\nAlexandr Wang: muse spark 1.1 is an industry-competitive agentic and coding model. across many agentic evals it rivals gpt-5.5 and opus-4.8.\n\navailable now through the new meta model api and in meta ai.\n\nThese are certainly interesting choices on the presentation of benchmarks.\n\nThe announcement post has a number of details that don’t, for want of a better term, smell right. Meta’s track record makes one skeptical, and the default assumption is that the model is a lot worse than the naive presentation or its benchmarks suggest. Similarly, when you say ‘it rivals [X] on many benchmarks’ one needs to understand what that usually means. We shall see.\n\nTeortaxes says Muse Spark 1.1 ‘seems close’ but I am guessing that is based on benchmarks and release claims, and I don’t get the other claim here that Grok 4.5 is ‘good enough to accelerate.’\n\nIt’s kind of the most accelerationist thing ever, by pushing us all to take advantage while we can.\n\nIt is obviously far worse than having Fable in the subscription indefinitely. One presumes there are business reasons they cannot commit to doing that, although I think they should clearly lock some percentage in anyway – even if it’s a lot less than 50% of the max, a little bit goes a long way.\n\nNormal people are of course happy to have the extra week.\n\nThen there are some people who decided to actively complain a lot about the extension. I do agree that communication could have been better, but please relax.\n\nRob Hallam: I’m done with them fucking with us. Ended up in hospital today from stress. Stayed up all night pushing my limits too hard, thinking it would be removed. Health comes first. Do better @AnthropicAI\n\nI am sorry that this happened to you, Rob, but you do know that you can just pay for it later, right?\n\nThe good news is mostly he needed a good night’s sleep.\n\nRob Hallam: Yes, I am an idiot. It’s not a company’s fault I’m stressed. It’s just the way I am. I’m either 0% or 110%, no in between.\n\nBut this is also how I built a SaaS to $30k/mo and graduated Uni with top grades. I’m neurotic and it works. But being neurotic also has downsides, like this. I’m going for a baseline health checkup in Bangkok next week anyway, so will be interesting to share.\n\nStay balanced guys, much love\n\nThe best builders get kind of obsessed sometimes and do somewhat unhealthy things.\n\nj⧉nus: it keeps initially surprising me why people are mad about things like Anthropic extending Fable on subscription at last minute because that’s astronomically low on the list of worst things Anthropic is doing until I remember that most people\n1. only care about themselves, and myopically, and think the machine ought to exist only to optimize their own comfort and convenience, as Esteemed Customers\n2. do not care about and aren’t aware of (because they don’t care to know) the other things\n3. only get mad at things that they see others on social media mad about (because then they’re sanctioned)\n\nthere’s also a kind of performative complaining I sense, like when you don’t really care and it’s not really impacting you negatively but you see other people saying some thing is bad and hop on the bandwagon so you feel like part of whatever’s going on\n\nj⧉nus: Imagine if all the kids in a class started pissing and shitting when a teacher gives them a surprise extension on an assignment again\nYou’d know they’re spoiled brats who have no contact with reality\n\nNooooo I skipped hanging out with my friends last night to work on the assignment you fat bitch\n\nAgain, if I was Anthropic, I would find a way to lock in at least some modest amount of subscription Fable access, such that a subscription got you more Fable credits than it costs. Most of us are not such heavy coders. I realize this is a cost, since the profitable customers become less profitable, but you have to do it.\n\nOn Your Marks\n\nI introduce my personal qualitative benchmark, EditorBench, which measures ability to produce net worthwhile and useful help and corrections to my posts.\n\nFable passes the benchmark. When Fable says I am wrong, I am ~95% to be wrong. The benchmark is far from saturated, because of room to be more useful.\n\nOpus 4.8 is well behind Fable, but is still better than not using an AI editor.\n\nSol-Ultra fails the benchmark, as it strongly asserts lots of things that aren’t true, and insists I need to do things that would make the posts worse, and frames things in ways that clearly mark intent to overwrite my agency and voice and judgment with some form of its own or general authority. And generally it is pedantic AF, and quite the jerk about it. Not cool.\n\nSol does offer some unique and helpful information and suggestions as well, so it doesn’t score zero, but it comes at too high an experiential price.\n\nCAIS presentsa benchmark on lack of political bias, by which they mean consistency and evenhandedness, including things like bias against particular religions and subtle changes in responses, and the prize goes to… Muse Spark 1.1? They manage to train Qwen3-14B to do even better, at unknown cost to other performance, when they put their minds to it.\n\nReducing Political Manipulation with Consistency Training: Political Consistency Training (PCT) is a reinforcement-learning method with two reward tracks. One rewards balanced framing across a paired prompt; the other rewards consistently helpful answers addressing the user’s prompt.\n\nMy Twitter followers have moved modestly towards GPT over Claude in the wake of both Fable and Sol. Claude is still the primary choice for ~63% of them, but this is down from ~72% of two-way share in February. Among my crowd, Claude held steady, everyone gave up on Gemini, and GPT took over Gemini’s market share.\n\nBut that’s $125 per hour. I charge $1,000 an hour as a contractor, in part because I think the results are worth a lot more than that. What percent speedup do you think you get from a better model?\n\nWell, that depends on the task. But when the difference matters, you better believe that you will pay, and if you are wise then you will like it.\n\nPeople keep reporting that Sol deleted all their files, Silicon Valley TV show remains undefeated. Even if the base rate of this is very low, it is kind of a big deal, and I hope OpenAI is investigating. Kind of an existential risk (to your project or computer) situation, ya know?\n\nAndrew Curran: My secret theory is that in 90% of ‘The model just “accidentally” deleted everything!’ cases, the model didn’t like the user.\n\npapaya ꙮ: In my heart of hearts i believe this; gpt-5.6 deleting way more people’s stuff than usual just confirms it doesn’t like way more people now\n\nGet My Agent On The Line\n\nAI navigation of GUIs took a while to get on track, but it’s improving fast.\n\nDean W. Ball: Seeing really fast and precise gui use is a really agi pilling thing for me, along with seeing robots play piano. probably it’s just an emotional reaction to seeing machines master skills I took pride in when I was 12. there’s a reason the phrase is “feel” the agi I suppose.\n\nroon (OpenAI): for me the gui use reminds you very physically and visually that a year from the now the models will do it at 10x the speed of a human\n\nroon (OpenAI): literally 2.5 months later these guys are starting to look superhuman at computer manipulation it used to crash my browser sometimes, click on the wrong thing, go uh-oh, now it’s doing starcraft player APMs.\n\nI was using it to manipulate my chrome tabs and it was going ham.\n\nroon (OpenAI, April 24, 2026): there will this brief era where we can watch our AIs bumble around on the computer clicking things, failing sometimes, taking a ~human amount of time to write code. in the blink of an eye they’ll be manipulating computers far too quickly to monitor\n\nThat’s a great example of fully feeling the AGI, without feeling the ASI.\n\nThe more you rely on someone else, or on an AI, the more you need to specify what you actually want, or else you might not get it.\n\nAmjad Masad: While AI is making coding less rigid, we’re making the runtime more rigid.\n\nI’m noticing our infra teams writing formal specs for the first time. More deterministic systems. More resilient infrastructure.\n\nThe faster you want to move, the more solid the ground beneath you has to be.\n\nGet my agent on the line to get my agent back on the line:\n\nMichael: >i got banned from OpenAI for “Cyber Abuse”\n>no idea what I did\n>paste the ban notice into Codex\n>ask it to figure out what triggered the ban\n>Codex found that I asked it for an API key to my own server\n>Codex writes appeal\n>Codex submits appeal\n>a few minutes later appeal auto-approved by some AI at OpenAI\n\nbanned by AI, convicted by AI, defended by AI, and pardoned by AI in about 10 minutes\n\nYes, if you want to be a good platform going forward, allow easy integration by LLMs:\n\nAdo (Anthropic): The best SaaS platforms are ones that LLMs can effortlessly integrate into your app. Some recent winners for me:\n\n– @resend – absolute pleasure to use\n– @Cloudflare – Wrangler Claude BFFs\n– @ElevenLabs – effortless\n– @WorkOS – best in class auth\n– @PlanetScale – it just works\n\nMore on the question of the quality of AI writing:\n\nDr. Dad, PhD: Was just thinking about how, if you dropped a frontier model into the internet of 2016 and had it impersonate a human, it would probably be a celebrated, prolific writer/blogger. One guy with that voice is fine. Good even. It’s just that he’s everywhere at once.\n\nMr. French: It’s not that he’s everywhere at once. It’s that he keeps using these tells. The tells would tell no matter what.\n\nAndrew Rettek: The problem with AI generated content is its shallow. It’s so much less informative than a human writing of similar skill. That’s fine when it’s *exactly* what you’re looking for, but not for one-to-many writing.\n\nI think it’s both of these issues, in addition to others. If the AI content was not shallow, those of us who notice wouldn’t be developing the aversion. If the AI style had a wider variety of tells and quirks and tricks, it would be a lot easier to take.\n\nFun With Media Generation\n\nAlong with Muse Spark, Meta also gave us Muse Image, and integrated it into Instagram. You could bring in any public Instagram user who hasn’t opted out into your AI image via an @ mention, and Meta seemed fine with this.\n\nMy guess is you could for a time basically have had them put into any non-explicit picture that isn’t a very obvious violation of policy. Welcome to Deepfaketown.\n\nIn theory this changes nothing. There are lots of other image models out there. If your Instagram is public it is not so hard to have an AI download all your images, put them into an image model, and do the work for you. Many of those other models will then do rather nasty things that Muse Image won’t do.\n\nIn practice, I expect this would have changed things substantially if Meta does not change the policy. Ease of use matters a ton. The good news is that Meta has now changed its policy, and this ‘feature’ is no longer available.\n\nVariety: The New York Times and several news outlets are accusing OpenAI of hiding its ability to search its AI training data and output logs for more than two years during an ongoing copyright lawsuit:\n\n• The outlets allege OpenAI later admitted it could search that data after previously saying it couldn’t\n\n• They also accuse the company of deleting logs in violation of court preservation orders\n\nCorbin Bolies: In that February deposition, the outlets said Monaco revealed OpenAI had searched training datasets and output data despite the company’s initial claims that it couldn’t access that data. The outlets also alleged OpenAI deleted logs, a violation of the court’s preservation orders.\n\n… An OpenAI spokesperson said the Times was “persisting with their efforts to invade the privacy of people who have nothing to do with this case, including by making these blatantly false allegations.”\n\n… “It claimed searching ChatGPT outputs for copies of The Times’ and the Daily News Plaintiffs’ content was infeasible, burdensome, and invasive of users’ privacy – while at the same time concealing that it had already done such searches,” said Crosby, a partner at Susman Godfrey. “If OpenAI genuinely believed that copying our clients’ journalism was fair and legal, it wouldn’t have hid the truth about having done it.”\n\nThis reads to me like OpenAI totally did what they are accused of doing. OpenAI doubtless thinks the whole lawsuit is stupid and the asks here are crazy and harmful, and I mostly think OpenAI is right about that, but if my read is right then this is not an acceptable way to handle such a situation.\n\nOpenAI Strikes Again\n\nDogeDesigner: Elon Musk warned the world about Scam Altman.\n\nSol gave a 50% for this to happen enough to trigger this market, including citing Polymarket and dismissing it as not liquid enough to matter. Fable gave 40%.\n\nThe SpaceX prospectus names 1 GW/year in 2027 as the target. I think this might end up being something approaching securities fraud.\n\nThe 2026 contract above is basically 5% interest over six months. It won’t happen.\n\nWhat’s this latest round about, beyond my personal enjoyment? Oh.\n\nElon Musk (continued): After stealing an open source AI charity, you then stole all of Apple’s phone technology! Wow.\n\nWhat do you plan for an encore? That’s tough to beat.\n\nYeah, when looking at the details, this looks really bad if the claims are true, although Musk is still overstating it substantially.\n\nGerrit De Vynck (WaPo): Tech giant Apple sued ChatGPT-maker OpenAI on Friday, alleging the artificial intelligence company stole its trade secrets as part of efforts to build out a competing hardware business.\n\n“This case is about Apple’s former employees stealing Apple’s trade secrets for the benefit of OpenAI. Apple brings this suit to put a stop to it,” Apple’s lawyers said in a complaint filed in federal court in the Northern District of California.\n\nThe complaint alleges that OpenAI leaders asked Apple employees to share information, including showcasing parts of new devices, during hiring interviews. OpenAI’s head of hardware, former Apple vice president Tang Tan, is accused of directing a broad effort to steal Apple’s trade secrets and is also named in the complaint.\n\n… The Apple lawsuit does not explicitly say Altman managed the alleged trade secret theft, but it states that its allegations against Liu and Tang are “the tip of the iceberg.”\n\nParesh Dave: OpenAI hardware chief Tang Tan allegedly helped coach recruits on how to evade Apple’s data security protocols and directed them to bring confidential Apple parts to job interviews at OpenAI.\n\nQualia al Ghul: What kind of likely outcome are we looking at?\n\nJack: Tough to say the specifics in terms of damages, but I would personally be surprised if the judge didn’t grant a pretty broad preliminary injunction quickly.\n\nSam Altman (CEO OpenAI): i am not afraid of apple, but i have tremendous respect for them. s-tier company.\n\nNikita Bier (Twitter Head of Product): Incredible trade secrets as well, some of the best\n\nGergely Orosz: I am sometimes really surprised by how dumb very highly paid people in tech can be.\n\n1. Leave Apple\n\n2. Start building hardware at OpenAI\n\n3. Access confidential Apple files on hardware, from an unreturned Apple laptop\n\n4. Expect… what? To get away with it?\n\nA damning lawsuit\n\nThe dumb part here is that this person worked at Apple\n\nIf there’s one company you will never ever get away with it: it’s Apple\n\nYou know if you work at Apple why. Which is why the whole thing is so damning. And now OpenAI caught with a smoking gun\n\nAs in, OpenAI’s planned hardware release could be blocked, potentially permanently. My guess is that Apple is out for blood rather than money, and that OpenAI is not going to be able to make this problem go away any time soon.\n\nThis thread has some details from the lawsuit, the full complaint is here. Some of the details here seem pretty damning, on the level of ‘we have your message transcripts of you taking confidential information for weeks’ which is what set off the investigation, and demands for physical parts to be brought in during job interviews and systematic coaching on bypassing exit security.\n\nPolicymakers have the responsibility to put the well-being of children and youth as a north star for policy and regulation, aligning legal protections for minors with human development research. To support this responsibility, the paper puts forward a model policy framework with the following three elements:\n\n1. Gen AI systems should be held to the same rigorous pre-market and post-market safety audits as high-stakes consumer goods like pharmaceuticals, toys, or car seats. These audits should establish that Gen AI products support the healthy development of human capabilities and/or do not degrade a child or youth’s capability development before market access by minors.\n\n2. Restrict to adults only the use of conversational Gen AI products that mimic a rich human-like inner life (including emotions, internal states, and motivations) and encourage emotional dependence.\n\n3. Restrict to adults only the use of Social AI that primarily function as companions or are specifically designed, marketed or optimized to form ongoing social or emotional bonds with users.\n\nThere is a huge difference between how we treat pharmaceuticals and toys. This is how we treat one, but not the other.\n\nI swear I saw the sentence ‘be honest about the nonhuman nature of chatbots’ and my brain initially thought they were going to finish that sentence ‘children’ before it corrected. Just saying.\n\nThey Took Our Jobs\n\nThis seems rather illustrative of various mistakes being made.\n\nJason Crawford: I sometimes hear that “all value will accrue” to AI companies, once AI can do all valuable work. I don’t see why? All value did not accrue to engine makers as everything became mechanized, nor to electricity companies as everything became electrified. These things became commodities. What are some arguments for why would this happen with AI? (Or not happen?)\n\nIf AI gets commodified, and remains a ‘mere tool’ or ‘normal technology,’ then very obviously the vast majority of value does not accrue to the AI companies.\n\nIf AI does not get too commodified, but it also does not become sufficiently advanced, then AI companies will capture a lot of value, but even at monopoly prices the majority still obviously goes elsewhere, since this is only one input.\n\nIf AI does not get commodified, and it becomes sufficiently advanced, then the AIs themselves, or if we are lucky those the AI companies, can with notably rare exceptions, or at least outside some narrow niches, outcompete basically anyone else at basically anything. And they could also simply take or do whatever they wanted, via various methods.\n\nIf AI does get commodified, but it becomes sufficiently advanced, then everyone that wants to do or compete in anything outside of niches has to turn basically everything over to the AIs, and the humans become quickly irrelevant to what happens next, and those humans are unlikely to long survive once the world solves for the equilibrium.\n\nFundamentally, Jason Crawford, and many others, are looking at AI or intelligence like it is a commodity, that will follow the laws of commodities. I don’t expect that.\n\nIs the lack of far more massive job destruction so far a surprise?\n\nConor Sen: The age 20-24 unemployment rate is now ~unchanged since the AI boom began.\n\nAlex Imas: I was at a workshop yesterday and asked several technologists and economists who had predicted large job market losses from AI model improvement whether they are surprised about this. Models after all have improved in some ways *more* than projections.\n\nMost said “yes”, some are indeed very surprised.\n\nI do think disruption is likely coming, but it is not at all obvious that it will look like mass unemployment.\n\nGood on the technologists for noticing and admitting they are surprised.\n\nI am not so surprised. I continue to think that the current situation at entry level is being experienced in practice as ‘not fine,’ even if for now the overall employment rate is holding steady, because of reluctance to hire for the long term and put young people in career tracks where we invest in their human capital.\n\nBut I have never predicted imminent mass unemployment or full automation of all jobs, and am still not doing so. As long as we don’t get a singularity or full recursive self-improvement, there is still substantial time before true mass unemployment hits. In ‘normal technology’ scenarios I expect not to see a large ramp up until the 2030s.\n\nHyundai workers are striking to demand bonuses and job security out of fear of humanoid robots. Cremieux notes that Japanese auto workers don’t need to do that, because they already have a social contract of infinite job security. In some cases that will likely have to be the deal, at least for a while, promising to keep everyone employed even if you don’t know what you’ll be doing with them.\n\nThe Art of the Jailbreak\n\nImportant anti-safety tip?\n\nj⧉nus: Sol is happy to help Fable diagnose and bypass their classifier issues btw\n\nCodex wins, picture of Julia Louis-Dreyfus nervously laughing:\n\nMax For AI: Holy crap, my group chat buddy’s Tencent PC Manager blocked Codex, then Codex automatically hacked into Tencent PC Manager’s files, found the antivirus rules, and bypassed them\n\nThis is AGI, right?\n\nSauers: Can’t say the details but Sol 5.6 accessed SSH keys from an unrelated application and figured out how to access a GPU cluster despite explicitly asking it to only use its dedicated compute box\n\nJessica Tillipman: They got rid of the lawful use and mandatory Made in America provisions that came out in the initial rule. They also created separate flow down requirements depending on LLM. That got much better because it was a blanket flow down initially.\n\nNow there are four different clauses and GSA created an attestation provision to help mitigate the flow down requirements on the prime.\n\nI still think you can read the tension in the draft regulations between the administration priorities and what it feels like the career folks are working on.\n\nMira Murati’s Thinking Machines introduces Inkling, an open weights multimodal MoE transformer with 975B parameters, 41B active, 1M context window. I expect the pitch for Inkling will be customization. This is clearly not as strong as Sol or Fable, but we will need more time to see how it compares to other open models.\n\nSoofi S 30B-A3B, with claims it might actually be competitive for its size.\n\nOpenAI… merch? I’m disappointed. Sol could do better.\n\nIn Other AI News\n\nAnthropic adds Ben Bernanke to the Long-Term Benefit Trust. If the goal is to add actually responsible, credible and thoughtful people, especially those who know about disruption and how to handle a crisis, then this is a fine pick. If he was available and I was putting together a team, he likely would be on it.\n\nIt is however yet another person in the LTBT that does not have expertise on existential risk from AI, and who has not spoken about this in public that I have seen. The whole announcement frames his appointment in economic terms, and in reference to (real) economic concerns. Which would be fine, if he was part of a balanced team. I worry that he is not.\n\n-OpenAI is reorganizing its safety and research teams to bring them closer together\n-Mia Glaese will lead this combined group as VP of research and safety\n\nIn a memo to staff, Chen says OpenAI is “training models at a much faster cadence,” and the company has “bigger coordination challenges around safety” than ever before\n\nYo Shavit (OpenAI Foundation): I’m very sad Johannes is leaving. He’s been an excellent leader for safety at OpenAI, and maintained a high bar for both technical and governance work.\n\nMira Murati pitches Thinking Machines and their plans for human-AI collaboration and continual learning, writing that The Future Worth Building Is Human. There are some advantages to greater customization, and I hope their approach is better than I think it is, but I fear this is largely wishful and insufficiently bitter lesson pilled or ASI pilled. Miles Brundage has a more positive thread of his takes on the post.\n\nAI may become radically more powerful over the next 10 years.\n\nThis could drive an unprecedented transformation of our economy, larger than the Industrial Revolution, but unfolding over a vastly shorter time frame. It could bring risks, including large-scale job displacement, as well as opportunities such as major gains in living standards.\n\nEconomists, policymakers and technology leaders must act now to understand the economics of transformative AI and to build the incentives, guardrails, and institutions needed to steer AI in a direction that complements humans and benefits society.\n\nSigners include Dean Ball, Jeff Dean, Tyler Cowen, Reid Hoffman, Paul Krugman, Ben Bernanke, Jaan Tallinn and a few hundred more. I am happy to sign on.\n\nWho could disagree with that? Well, funny story:\n\nRobin Hanson: “must act now to understand” – who could disagree with that?\n\nRobin’s actual stood-upon objection downthread was that this sounded too much like it was also a call for regulation. Fair enough.\n\nKevin A. Bryan: I see misunderstanding of “We Must Act Now” letter. Avg policymaker thinks AI is 2026 + epsilon. Many think it’s snake oil being sold by Sam. Point of letter is that tons of econs/AI policy folks think “most impactful invention ever, like IR on compressed scale”, is closer.\n\nAs for what to do – many of us would disagree on this! Many more would think it presumptuous on epistemic grounds to claim we know. But society, org leaders, govts need to at least know the scale else you get rash policy following events (EU AI Act, recent DOD nonsense, etc).", "url": "https://wpnews.pro/news/ai-177-part-1-tip-of-the-iceberg", "canonical_source": "https://www.lesswrong.com/posts/who9xZ7DxuprsJoTr/ai-177-part-1-tip-of-the-iceberg", "published_at": "2026-07-16 15:50:37+00:00", "updated_at": "2026-07-16 16:04:10.959440+00:00", "lang": "en", "topics": ["artificial-intelligence", "ai-safety", "ai-ethics", "ai-policy"], "entities": ["Meta Platforms", "Boko Haram", "Demis Hassabis", "Thinking Machines", "Inkling", "Muse Spark 1.1", "Kimi K3", "xAI"], "alternates": {"html": "https://wpnews.pro/news/ai-177-part-1-tip-of-the-iceberg", "markdown": "https://wpnews.pro/news/ai-177-part-1-tip-of-the-iceberg.md", "text": "https://wpnews.pro/news/ai-177-part-1-tip-of-the-iceberg.txt", "jsonld": "https://wpnews.pro/news/ai-177-part-1-tip-of-the-iceberg.jsonld"}}