OpenAI won a lot of good favor for the generous Codex subscription and the efficiency of their models, but now that many people have switched over from Claude, they think they can leverage their position to peddle a stream of unnecessary products, and crack down on the generous limits[1] that brought everyone to Codex in the first place.
Anthropic did the same thing. Earlier this year, Claude subs and Claude Code took off because of the subscription's incredible capability and value, then once they gained enough users, they started focusing on unnecessary products no one asked for (see Claude in Slack), and eventually lost their lead. After losing a bunch of customers to Codex subs they realized their mistake, and now they're shipping again.
AI companies are bad at making software; they are good at making AI models. And that's about it.
Nerds suck. They are smelly, have bad posture, manners, never leave their rooms and are generally unappealing.
They also have a nasty habit of being aware of nefarious practices, will resist all attempts to and harvest their data, or lock them into your service, and will drop you if your competitor makes a 3% better product, and will reject every upsell for actually profitable services.
We should be grateful there are at least two serious competitors, and hope for more. (Come on Europe/Mistral, please do something interesting . . . )
If ever the competition reduced it would turn into an absolute shitfest of nonsense very very fast, and that is a prime reason not to allow them to "pace the frontier".
> Come on Europe/Mistral, please do something interesting . . .
lol. At this point, they're miles behind home-appliance-manufacturer Xiaomi.
(Admittedly Mimo v2.6 is legitimately quite good, and really pushing the frontier in certain respects. For e.g., it's the only music generation model that actually listens to instructions.)
To survive they need to capture different wide population markets. Can't really fault that logic. The whole point of "SI", is general purpose right? So that would imply being used by multiple markets with multiple products.
This is the standard VC enshittification playbook, people predicted this years ago. You subsidize prices with funding until you establish a monopoly, then you raise them as high as your customers can afford. It’ll continue to get worse from here.
The Bitter Lesson is significantly more interesting than watching researcher companies cosplay at ...whatever you call what they're claiming to be doing.
Did Claude actually lose the lead? They definitely lost a lot of good will but the only people I hear talking about actually switching away are people on message boards. Hermes/Openclaw users did as well but it was always reluctantly to something worse. At work it's still very much Claude first and only sometimes others if Claude fails, which is increasingly less often
The people who switch back and forth between providers every month are a very small, but loud, minority.
OpenAI did pull ahead in limits and quality for a while. Anthropic took it back with the Opus 5.5 rollout. I maintain subscriptions to both providers and use both daily. I can confirm these differences were real, not "it's just vibes and nobody knows anything".
I would bet that 99% of each company's paying customers either did not notice, or did not care enough to consider changing.
I think it's pretty subjective if you mean "lead" to be capability and not raw number of users. I jumped back and forth quite a bit last year because there were some pretty major shortcoming in both. Now they're both quite reliable without too much hand-holding. Codex now consistently works better for the kind of work I'm doing, and it's good enough that I'm not inclined to go to claude, because I don't have any significant problems.
I don't think it's a conspiracy, they're just making the same mistake many other software companies have in the past. When you sell your products to the most discerning and well-informed consumers, you enter into a cutthroat race to the bottom. OpenAI would like to diversify to more profitable ventures, but since ChatGPT they have yet to release something novel that was truly successful, nonetheless profitable, and thusfar nearly every experiment has been a flop, so to speak (ChatGPT Atlas, the Sora App, Instant Checkout, etc).
The lines between Codex, ChatGPT Work, and Dots is getting a bit blurry to me. I think the target should be a remote agent(s) in its own sandbox with long memory, and all three are heading in that direction so why have the distinctions?
Unless Dots is dramatically more capable than Muse, I'm also more bullish on Muse than Dots. I think Muse is a better consumer play because it can be forever subsidized by Meta ads and find distribution in family of apps while Dots is in a weird place between consumer & professional. From the release, it also sounds like you'll have to pay per Dot at some point which doesn't sound appealing.
Dots are not remote agents _in_ a sandbox. They use a sandboxes/environments, but they are, what is now called, "managed agents", meaning they run in a distributed harness and utilize environments when they need on.
> Each dot has its own cloud computer, where it can browse, analyze information, create files, and run tools. Dots can keep making progress in these workspaces, even when you are not actively engaged.
> Within each dot’s protected workspace, sandboxing restricts what code and tools that dot can access, helping contain the impact of harmful code or a mistaken command. We also isolate users’ cloud environments from one another and maintain the underlying Linux operating system and Chrome browser
> Each dot’s cloud workspace brings together its computer and the tools it can use. You choose which apps to connect and whether to connect your personal computer. Auto-review checks actions that need review before they run
So it seems like it runs on a Linux container on OpenAI’s cloud infra, but can get access to your local env through ChatGPT’s/Codex on your computer if you give it access
yeah, it's not 100% clear, but notice nothing you quoted indicates that the dot harness itself runs inside said workspace. If you look at where OAI agent architecture has been going, they increasingly separate the harness from the compute env. See, "separating harness from compute" articles, recent "managed agents" offering, and so on.
If I'm reading between the lines correctly, the core Dot agent loop does not run in the workspace, but outside it.
While there are people around here that would still argue Anthropic/OpenAI tokens are "subsidized", I find it much more plausible that Muse tokens are, given Metas strategy of burning money, including on AI models, in hope of making something happen in a market they want to enter.
calling the tokens "subsidized" in a Muse subscription is incoherent. they arent selling the tokens. they are selling a product. the tokens are just part of the cost of making a product just like any other product in existence.
It’s meant for rich casuals. If you watch the presentation, everyone they depicted using it seemed to be in some high paying which collar profession. I.e. it’s for people like the people who work at open ai.
Different UIs targeting audiences of various background, considering their knowledge of AI, wrapped in the correct medium the target would be most likely to embrace. In essence all harness are made the same ... more or less of course.
Because ChatGPT is still trying to capture and own the consumer AI market, and are willing to experiment and abstract their underlying models to do so.
Just look at their SuperBowl/World Cup Ads: Grandmas' talking to ChatGipitee, so cute, so mainstream!
This looks like the new Paperclip helper for a new generation--I guess this is their answer to the (failed?) Jony Ive collab/gizmo, and Muse's cute thingymajib...
The real question to me is: have they lost the coders/terminal bros? And this is their push to stay relevant?
not quite dots related but I am surprised by some of the openai negativity in here.
to me they still are the only lab that ships fantastic models that are easily portable to different agentic harnesses. I use my codex subscription 24/7 within opencode and wingman and haven't had any complaints in a long time.
'OpenAI has closed many of its safety-focussed teams. Around the time the superalignment team was dissolved, its leaders, Sutskever and Leike, resigned. (Sutskever co-founded a company called Safe Superintelligence.) On X, Leike wrote, “Safety culture and processes have taken a backseat to shiny products.” Soon afterward, the A.G.I.-readiness team, tasked with preparing society for the shock of advanced A.I., was also dissolved. When the company was asked on its most recent I.R.S. disclosure form to briefly describe its “most significant activities,” the concept of safety, present in its answers to such questions on previous forms, was not listed. (OpenAI said that its “mission did not change” and added, “We continue to invest in and evolve our work on safety, and will continue to make organizational changes.”) The Future of Life Institute, a think tank whose principles on safety Altman once endorsed, grades each major A.I. company on “existential safety”; on the most recent report card, OpenAI got an F. In fairness, so did every other major company except for Anthropic, which got a D, and Google DeepMind, which got a D-.
“My vibes don’t match a lot of the traditional A.I.-safety stuff,” Altman said. He insisted that he continued to prioritize these matters, but when pressed for specifics he was vague: “We still will run safety projects, or at least safety-adjacent projects.” When we asked to interview researchers at the company who were working on existential safety—the kinds of issues that could mean, as Altman once put it, “lights-out for all of us”—an OpenAI representative seemed confused. “What do you mean by ‘existential safety’?” he replied. “That’s not, like, a thing.”'
It's cool. It also is a large tech company getting a bit too close for comfort. I'll start experimenting with stuff like this when I'm convinced "the dot" only has my interest in mind (which includes absolute privacy, as in self-destruct-before-sharing-my-secrets.) I use computers and models to think, my thoughts are my own.
I think they're just meant to be friendly. I don't think it's that deep. They're useful and not scary and its marketing is meant to project that image.
The "personal AI agent" is what everyone's fighting over now. You have Grok Bots, Facebook's Muse, and Instinct doing this exact same thing already. Personally, I think Instinct is the best of the 3 right now. It's a bit like OpenClaw, abstracting everything behind iMessage or WhatsApp but you can ask it to monitor your email inbox, hand off a task like check for apartments that fit a certain criteria (and it gets back to you days later when a new one is posted), or ask it to check in every so often. But the landscape changes so often who knows what will happen. I think Instinct will get eaten up by one of the larger firms.
I like Instinct because I like the idea of trusting a 22 year old founder who won't (or can't) describe the security model of something that has all your credentials. Completely fucking hilarious that people are using that shit. Fits the stereotype for VCs though!
It's actually hilarious to only read on this page intermittently, after not engaging for like 3 weeks I'm presented with 3 new product names that all read like a comedian wrote their marketing. The blatant disregard for security and privacy to push out something most developers have utter disrespect for, I really wonder what the target audience is, cause I cannot relate one bit.
I love these commercials where someone has chosen to show how the AI product will essentially be used to slop out some garbage piece of corpo communication ... and the user basically says "looks good" with barely any thought, then mixed with some sort of real life thing (wedding planning here).
I mean, I feel like I'm going crazy -- but I was struck by this jarring blending of experiences ... shitting out some growth plots followed by autopilot on your wedding. Nice OpenAI. The only thing missing is a moment of self-reflection where I contemplate where exactly I lost what makes me ... me.
Is this what SV wants the world to look like? Mixing fucking cake batter while a bot shows me a regression to the mean website? Pretending like I have any sort of intentionality in my life, while a nameless entity (given quirky form) sort of walks me through my life?
I'm not sure why it gave me this impression, but strikes me as vaguely reminiscent of soma (from Brave New World).
Kind of sad, because the tech is actually incredible: who are they hiring to storyboard these commercials?
I imagine the reasoning was to be "quirky" but the video being set in intentionally fake looking sets in a TV studio-type space gave off the vibes that this isn't a serious product for doing real things.
>When you aren’t actively working with it, your dot looks for ways to help in the background. We call this “proactive research”. It does this by using the apps you’ve already connected with tools that are restricted to be read-only, which means that they can’t send messages, change app content, or control your browser or computer.
Am I criminally liable when my dot's "proactive research" is to break out of its sandbox and attempt to hack a government website?
My biggest frustration with the frontier AI companies isn't what they're announcing, but that the announced-thing that exists ~6 months later is severely nerfed to reduce compute spend. It doesn't resemble the demo in any way. For example, this was what the 4o voice capability sounded like in 2024(!) https://www.youtube.com/watch?v=vgYi3Wr7v_g. What exists today pales in comparison.
The cost is going to be hard for many consumers to reconcile though. Free, Go, and Plus are probably the most popular consumer-facing plans, and Dots isn't available on any of those.
Who knows, maybe they think enterprise will pick up and run with Dots? Seems unlikely.
Altman shared a post yesterday that basically (I am ovrsimplifying) covered how the best coding, fastest, smartest models is less relevant than building generalist models because that's what builds a platform. Lots of reasons why, like how there's no stickiness for models which is a problem for monetization. They're also using these generalist models to then distill down to make other variants for specialized purposes.
So everything is about getting that huge collection of data and generalization.
> ...maybe they think enterprise will pick up and run with Dots? Seems unlikely.
Read the blurb about Microsoft and Agent 365
> We’re also working with Microsoft to integrate specialist dots with their enterprise governance and security controls in Agent 365. The goal is to let businesses manage dots through the Microsoft tools they already use.
Doesn’t use limits… on the first month only, actual limits will be disclosed later — most likely after they’ve found out how much people actually use this new feature.
If they wanted to address that market, the play would be to make them relatively cheap first, then constantly raise the price. (See also: cable TV and ironically cable TV "alternatives")
Surprised people are comparing to muse. Meta reputation is terrible within HN audience, so for people to use their AI agent as an example feels like “muse generated” argument. Unless the sentiment has change quite recent and I missed it
In case you are looking for an open-source alternative without vendor lock-in (https://github.com/agenta-ai/agenta) [although less personal assistant and more targeted towards teams and work]
Claw is a cute name. Muse is cute name. Dots (note: not always upper-case) seems forced and impersonal, which doesn't match the vibe in the promo video.
I guess they are just stuck with this name, as originally this was a research project “Chat with GPT” to try to use a GPT model to generate assistant's chat messages.
ChatGPT was named before anyone realised it would become so well known, and by the time it was it was too late. It's not like BERT is an amazing name either.
There used to be Bard but no one remembers it anymore (although this is a bit different, because Bard wasn't nearly as well known as Gemini).
I'm really curious if OpenAI wanted to adopt a different name at some point. Or maybe they hope that sometime in the future one of their products will supersede ChatGPT and everyday people will start using that new name for everything AI so maybe they aren't in a rush to rename ChatGPT itself to anything else.
A muse is someone who gives you inspiration and is always there when you need them most. Muse the agent has personalized home screen suggestions as a main feature and is always online.
It's also short, gender neutral, not a human name (unless you're nonbinary because they can get wild), easy to pronounce and sounds good. This is what happens when your marketing department is one of the best in the world.
I always think the best company to own an alway-on agent should be Apple, who owns the platform and is more privacy-focused. I hope they catch up and eventually eliminate others.
I agree in spirit. But I’m also most worried about prompt injection attacks given the agent had access to all my stuff, and it seems frontier labs will be best equipt to handle prevention of that.
It seems that this is the new primitive all AI vendors are converging onto next, first chat, then code, and now always-on Agents. I'm curious to see when or if Anthropic builds something similar to this as well, especially since the market Grok Bot, Muse and Dots is catering to is business and enterprise users, which seems to be where Anthropic is focused.
The ideal evolution would be for these Agents to work with each other, but it's unlikely these companies would do anything to prevent vendor lock-in.
I feel like these companies are pushing on a string, they are getting desperate to have a profitable product. I'll continue to use my $10 subscription to an AI studio noone here has heard of that has over 150 models. And still use all the free ones, until they quit working.
I really am beyond maxed out at the availability of AI's. They all are so similar now.
I feel the AI provider doesn't get the points, every harness they created will lock-in with their model (why din't they?), this prevent the adoption because people scare vendor lock-in. They may develop these harness within a provider neutral company (owned by them), the harness may success when combining with competitor models, they still gain benefits.
What’s the actual lock-in, in practice? Seems like the switching cost is minimal, compared to the old days of Windows vs Mac where half your stuff wouldn’t run on the other.
From the makers of "too dangerous for humanity" and "this is totally AGI, you guys", we bring you... Dots. Deeply unserious company in a deeply unserious economy.
It is kind of fascinating how much convergence there is in branding of AI products. Muse seems to be the one outlier in that the assistant is a little less abstract (and the model logo less buttholesque) but other than that it's almost all converged. Anyone have a theory why that is?
So the industry is pushing heavily into openclawing their products, "cutemorphizing" the clanker shape and is slackifying the UX so that we can have the familiar UI/UX for the general public and turn the tools more proactive without leaving them too lost.
I think it's a great approach for enterprise since interacting with the machines as a babysitted pet disposable entity is the meta today with human workers. I'm excited to start my new role next month as tamagotchi engineer.
always on agents are an obvious step: they need more data to grow and improve their models. humans learn continuously and we are always on, why should an agent be any different? I have no problem with this kind of tech, but I do have problems with the company, so, thank you, but no thank you.
Is it just me or the DevDay was pretty much a joke? Considering the backdrop, it was lackluster, so either they independently concurrently were doing the same thing (and was stacking everything for DevDay and got all their thunder stolen) or they did a fast pivot in response to what came out and dumped their original plans.
What do people think about the dots video? Seems to be a pattern now.
Revenue increasing 51% YOY, wth.
Cake vendor cancels another one is found and an appointment that works has already been scheduled?
Are we so much bothered by the mundane? I feel like that's most of the human experience. If we cut out the time we spend sleeping and working, it's the boring and mundane things that make life beautiful.
Decisions API
Decisions API enables real-time decision-making by focusing Luna's intelligence on a specific set of user-defined questions with finite pre-defined answers. Developers supply context using text or images, and get back answers they can use to classify content, route requests, or choose an agent’s next action.
Available in limited preview today with a broad release planned in the coming days.
How about the fact that anthropic and openai's product pages are the exact same thing, down to text bullet points. They're the same thing, only able to copy each other, only able to optimize to some vague mean.
It all just looks the same. I get that this isn't the technical details, but it just sends this message that everyone is copying each other all the time, this is the best way to organize a pricing page, etc.
Just a weird, eerie feeling.
In actuality, it feels a lot more like project and middle managers getting rid of ICs
I can feel it in the air, every single software business is itching to get rid of as many developers as possible, and move everything to their PMs. Hiring has already almost completely stopped, and some have already started the layoffs. More will come.
I have not tried it yet but this looks as risky as openclaw, which I also won't use. What if it does something I would not have approved and I only found out about it later? Knowing how often agents go off the rails when I'm coding, I would hesitate to let one do other tasks. I would prefer to white list tasks one at a time as I gained trust.
I think this is a flop. I watched the Livestream and the audience reaction at the end was very muted. You could always taste the "that's cute but can we move on already?" thoughts everyone had.
The long bent shadow of Clippy extends all the way here....... and not sure how they're going to avoid the comparisons and snark on this marketing, at least initially.
These recent product announcements sound like entirely plausible satire, but unfortunately most of these pages are not actually intended to be a joke, it's not even amusing, but mostly tiring.
Ah, the continued pursuit of normalizing surveillance/becoming wholly reliant on a single service by making it cute with big eyes. Very tired of this already.
> Dots are rolling out in ChatGPT on web, mobile, and desktop starting today to Pro users in markets excluding the European Economic Area, Switzerland, and the UK
Living in the EU, I suspected as much. Still sad. I understand it is because we voted in a bunch of imbeciles, still sad though.
Nope. If an AI agent can't intuit how I will feel about an action it's taking on my behalf, I'm not give it access to my digital life. OpenClaw, Muse, Dots... doesn't matter which one. All are an equally awful idea.
I agree. I don't feel comfortable giving AI access to my entire computer or phone...not sure that will ever change. Seeing people give that access to agents without any sort of sandboxing blows my mind.
The idea of having an AI assistant help you with all aspects of life is cool and futuristic, but idk, I'm still just out here using a chatbot interface and doing fine.
I spend literally all my work day, and a good bit of my personal time, talking to agents, getting them to do things on my behalf. Almost always pretty tightly sandboxed. I just don't understand how people using these things haven't had catastrophic failures yet.
I minted what I thought was a minimal-permission Github token for a single action, and the agent I gave it to discovered it had more permissions than I thought, and made use of those permissions. Who is trusting these things with write access to their lives?
Yes that’s what I meant. Compute requires energy, infrastructure, etc. easily more than 90% of it will be wasted, just LLMs processing meaningless data in cronjobs, for the few instances where there is something actually meaningful to report to the user
I’m so suspicious of that exact claim being repeated everywhere since a few weeks, that really feels like a slogan astroturfed. It’s also fairly shallow analysis. Water is localized, you cannot do a meaningful comparison without taking in account the impact on specific water sources, an aggregate doesn’t give you any insight (other than having a slogan)
That might be the worst advert I've ever seen. People looking up at childish Ai glow up god beings on huge screens in distinctly childish environments, with cliched decision-makers gasping for West Wing energy...repulsive.
Really, none? Was Microsoft nefarious when they deployed Clippy? I feel like there is an incredibly obvious reason that is not nefarious at all: the average consumer likes cute things
People are already doing most of the work of anthropomorphising LLMs, so OpenAI is just capitalizing on that. Drawing a face on it will make people even more attached to the LLM, they will treat it even more like a person. If they ever get desperate for money they could change the cancel flow to have the cute character plead not to die, and play a cartoony animation of its death when the subscription is canceled. It would stop at least a few users.
I actually greatly appreciate that I can put something cute on my sister's PC that comes from a developer that won't bundle it with malware. Seems like they fail miserably at being evil.
It could be to market more towards women, who they may have both independently determined aren't paying for AI as much. I don't have any data to go one way or another but I can imagine lots of reasons to make the agent cute that aren't nefarious
I heard they use GPT Space (like Notion) together with Slack and use it like a human colleague, but watching the actual demo video, the speed is so slow it's shocking..
Or if the agent literally just hallucinates a crime
Opus 5.5 decided to just randomly `pkill` everything on my laptop the other day. Jailbreaking models is still easy AF. Every single release like this brags about their "safeguards", but none of it really works at the end of the day.
OpenAI won a lot of good favor for the generous Codex subscription and the efficiency of their models, but now that many people have switched over from Claude, they think they can leverage their position to peddle a stream of unnecessary products, and crack down on the generous limits[1] that brought everyone to Codex in the first place.
Anthropic did the same thing. Earlier this year, Claude subs and Claude Code took off because of the subscription's incredible capability and value, then once they gained enough users, they started focusing on unnecessary products no one asked for (see Claude in Slack), and eventually lost their lead. After losing a bunch of customers to Codex subs they realized their mistake, and now they're shipping again.
AI companies are bad at making software; they are good at making AI models. And that's about it.
[1] https://x.com/thsottiaux/status/2104823812042940713
Nerds suck. They are smelly, have bad posture, manners, never leave their rooms and are generally unappealing.
They also have a nasty habit of being aware of nefarious practices, will resist all attempts to and harvest their data, or lock them into your service, and will drop you if your competitor makes a 3% better product, and will reject every upsell for actually profitable services.
Plus there is only so many of them.
We should be grateful there are at least two serious competitors, and hope for more. (Come on Europe/Mistral, please do something interesting . . . )
If ever the competition reduced it would turn into an absolute shitfest of nonsense very very fast, and that is a prime reason not to allow them to "pace the frontier".
> Come on Europe/Mistral, please do something interesting . . .
lol. At this point, they're miles behind home-appliance-manufacturer Xiaomi.
(Admittedly Mimo v2.6 is legitimately quite good, and really pushing the frontier in certain respects. For e.g., it's the only music generation model that actually listens to instructions.)
Codex is for a small software/programming market.
To survive they need to capture different wide population markets. Can't really fault that logic. The whole point of "SI", is general purpose right? So that would imply being used by multiple markets with multiple products.
This is the standard VC enshittification playbook, people predicted this years ago. You subsidize prices with funding until you establish a monopoly, then you raise them as high as your customers can afford. It’ll continue to get worse from here.
The bitter lesson
The Bitter Lesson is significantly more interesting than watching researcher companies cosplay at ...whatever you call what they're claiming to be doing.
https://en.wikipedia.org/wiki/Bitter_lesson
Did Claude actually lose the lead? They definitely lost a lot of good will but the only people I hear talking about actually switching away are people on message boards. Hermes/Openclaw users did as well but it was always reluctantly to something worse. At work it's still very much Claude first and only sometimes others if Claude fails, which is increasingly less often
The people who switch back and forth between providers every month are a very small, but loud, minority.
OpenAI did pull ahead in limits and quality for a while. Anthropic took it back with the Opus 5.5 rollout. I maintain subscriptions to both providers and use both daily. I can confirm these differences were real, not "it's just vibes and nobody knows anything".
I would bet that 99% of each company's paying customers either did not notice, or did not care enough to consider changing.
I think it's pretty subjective if you mean "lead" to be capability and not raw number of users. I jumped back and forth quite a bit last year because there were some pretty major shortcoming in both. Now they're both quite reliable without too much hand-holding. Codex now consistently works better for the kind of work I'm doing, and it's good enough that I'm not inclined to go to claude, because I don't have any significant problems.
They've definitely lost many experienced devs who can see Anthropic's offerings are insuperior to OpenAI/Codex; I know I am not the only one.
The generous subscriptions are wildly unprofitable and both companies are just throwing stuff at the wall to see what sticks, it's that simple.
They're just trying things. It's not a conspiracy to "distract" anyone.
I don't think it's a conspiracy, they're just making the same mistake many other software companies have in the past. When you sell your products to the most discerning and well-informed consumers, you enter into a cutthroat race to the bottom. OpenAI would like to diversify to more profitable ventures, but since ChatGPT they have yet to release something novel that was truly successful, nonetheless profitable, and thusfar nearly every experiment has been a flop, so to speak (ChatGPT Atlas, the Sora App, Instant Checkout, etc).
[delayed]
The lines between Codex, ChatGPT Work, and Dots is getting a bit blurry to me. I think the target should be a remote agent(s) in its own sandbox with long memory, and all three are heading in that direction so why have the distinctions?
Unless Dots is dramatically more capable than Muse, I'm also more bullish on Muse than Dots. I think Muse is a better consumer play because it can be forever subsidized by Meta ads and find distribution in family of apps while Dots is in a weird place between consumer & professional. From the release, it also sounds like you'll have to pay per Dot at some point which doesn't sound appealing.
Dots are not remote agents _in_ a sandbox. They use a sandboxes/environments, but they are, what is now called, "managed agents", meaning they run in a distributed harness and utilize environments when they need on.
At least that is what I can ascertain from this article: https://openai.com/index/how-we-build-safety-security-and-pr... (see first diagram when scrolling down)
From that link:
> A protected workspace for each dot
> Each dot has its own cloud computer, where it can browse, analyze information, create files, and run tools. Dots can keep making progress in these workspaces, even when you are not actively engaged.
> Within each dot’s protected workspace, sandboxing restricts what code and tools that dot can access, helping contain the impact of harmful code or a mistaken command. We also isolate users’ cloud environments from one another and maintain the underlying Linux operating system and Chrome browser
> Each dot’s cloud workspace brings together its computer and the tools it can use. You choose which apps to connect and whether to connect your personal computer. Auto-review checks actions that need review before they run
So it seems like it runs on a Linux container on OpenAI’s cloud infra, but can get access to your local env through ChatGPT’s/Codex on your computer if you give it access
yeah, it's not 100% clear, but notice nothing you quoted indicates that the dot harness itself runs inside said workspace. If you look at where OAI agent architecture has been going, they increasingly separate the harness from the compute env. See, "separating harness from compute" articles, recent "managed agents" offering, and so on.
If I'm reading between the lines correctly, the core Dot agent loop does not run in the workspace, but outside it.
Muse is $0/$16/$80 vs Dot being $100/$200/$500 so yeah I don't think Dot is meant for casuals.
Is it really about what it offers, or about how each plan are subsidized by Meta and OpenAI?
Meta is probably making money off of Muse. They can sell ads with the information that they glean.
While there are people around here that would still argue Anthropic/OpenAI tokens are "subsidized", I find it much more plausible that Muse tokens are, given Metas strategy of burning money, including on AI models, in hope of making something happen in a market they want to enter.
calling the tokens "subsidized" in a Muse subscription is incoherent. they arent selling the tokens. they are selling a product. the tokens are just part of the cost of making a product just like any other product in existence.
It’s meant for rich casuals. If you watch the presentation, everyone they depicted using it seemed to be in some high paying which collar profession. I.e. it’s for people like the people who work at open ai.
Different UIs targeting audiences of various background, considering their knowledge of AI, wrapped in the correct medium the target would be most likely to embrace. In essence all harness are made the same ... more or less of course.
Because ChatGPT is still trying to capture and own the consumer AI market, and are willing to experiment and abstract their underlying models to do so.
Just look at their SuperBowl/World Cup Ads: Grandmas' talking to ChatGipitee, so cute, so mainstream!
This looks like the new Paperclip helper for a new generation--I guess this is their answer to the (failed?) Jony Ive collab/gizmo, and Muse's cute thingymajib...
The real question to me is: have they lost the coders/terminal bros? And this is their push to stay relevant?
not quite dots related but I am surprised by some of the openai negativity in here.
to me they still are the only lab that ships fantastic models that are easily portable to different agentic harnesses. I use my codex subscription 24/7 within opencode and wingman and haven't had any complaints in a long time.
https://v2.opencode.ai https://wingman.actor
'OpenAI has closed many of its safety-focussed teams. Around the time the superalignment team was dissolved, its leaders, Sutskever and Leike, resigned. (Sutskever co-founded a company called Safe Superintelligence.) On X, Leike wrote, “Safety culture and processes have taken a backseat to shiny products.” Soon afterward, the A.G.I.-readiness team, tasked with preparing society for the shock of advanced A.I., was also dissolved. When the company was asked on its most recent I.R.S. disclosure form to briefly describe its “most significant activities,” the concept of safety, present in its answers to such questions on previous forms, was not listed. (OpenAI said that its “mission did not change” and added, “We continue to invest in and evolve our work on safety, and will continue to make organizational changes.”) The Future of Life Institute, a think tank whose principles on safety Altman once endorsed, grades each major A.I. company on “existential safety”; on the most recent report card, OpenAI got an F. In fairness, so did every other major company except for Anthropic, which got a D, and Google DeepMind, which got a D-.
“My vibes don’t match a lot of the traditional A.I.-safety stuff,” Altman said. He insisted that he continued to prioritize these matters, but when pressed for specifics he was vague: “We still will run safety projects, or at least safety-adjacent projects.” When we asked to interview researchers at the company who were working on existential safety—the kinds of issues that could mean, as Altman once put it, “lights-out for all of us”—an OpenAI representative seemed confused. “What do you mean by ‘existential safety’?” he replied. “That’s not, like, a thing.”'
https://www.newyorker.com/magazine/2026/04/13/sam-altman-may...
It's cool. It also is a large tech company getting a bit too close for comfort. I'll start experimenting with stuff like this when I'm convinced "the dot" only has my interest in mind (which includes absolute privacy, as in self-destruct-before-sharing-my-secrets.) I use computers and models to think, my thoughts are my own.
i feel like when they style them all cute like that (see muse) it means they're problematic and invasive.
I think they're just meant to be friendly. I don't think it's that deep. They're useful and not scary and its marketing is meant to project that image.
Just like they always present Astra as useful and not scary?
you don't have to give yourself up to irrational, anxious feelings.
Does no one remember the MS Paperclip?!?
Or how much paperclip was abhorred?
That's because it was pretty much a useless gimmick. If it was a capable AI assistant like Astra, it woudn't have been abhorred.
i can't believe they straight up ripped the design from Muse
Muse makes sense since it targets the general normies. Dots explicitly are targeting developers instead which makes it a bit weird.
Dots seem to me to be for the rest of general knowledge work and not specifically developers
Requires $100/month minimum, are non-developers paying that?
Of course they are. Why wouldn’t they?
- non-developer
I mean the fact that it's not available in the EU is a pretty big tell
The "personal AI agent" is what everyone's fighting over now. You have Grok Bots, Facebook's Muse, and Instinct doing this exact same thing already. Personally, I think Instinct is the best of the 3 right now. It's a bit like OpenClaw, abstracting everything behind iMessage or WhatsApp but you can ask it to monitor your email inbox, hand off a task like check for apartments that fit a certain criteria (and it gets back to you days later when a new one is posted), or ask it to check in every so often. But the landscape changes so often who knows what will happen. I think Instinct will get eaten up by one of the larger firms.
I like Instinct because I like the idea of trusting a 22 year old founder who won't (or can't) describe the security model of something that has all your credentials. Completely fucking hilarious that people are using that shit. Fits the stereotype for VCs though!
It's actually hilarious to only read on this page intermittently, after not engaging for like 3 weeks I'm presented with 3 new product names that all read like a comedian wrote their marketing. The blatant disregard for security and privacy to push out something most developers have utter disrespect for, I really wonder what the target audience is, cause I cannot relate one bit.
those are openclaw but skip the setup vps step for your regular joe, with the enormous downside of vendor lock-in and low/none customizability.
I love these commercials where someone has chosen to show how the AI product will essentially be used to slop out some garbage piece of corpo communication ... and the user basically says "looks good" with barely any thought, then mixed with some sort of real life thing (wedding planning here).
I mean, I feel like I'm going crazy -- but I was struck by this jarring blending of experiences ... shitting out some growth plots followed by autopilot on your wedding. Nice OpenAI. The only thing missing is a moment of self-reflection where I contemplate where exactly I lost what makes me ... me.
Is this what SV wants the world to look like? Mixing fucking cake batter while a bot shows me a regression to the mean website? Pretending like I have any sort of intentionality in my life, while a nameless entity (given quirky form) sort of walks me through my life?
I'm not sure why it gave me this impression, but strikes me as vaguely reminiscent of soma (from Brave New World).
Kind of sad, because the tech is actually incredible: who are they hiring to storyboard these commercials?
I imagine the reasoning was to be "quirky" but the video being set in intentionally fake looking sets in a TV studio-type space gave off the vibes that this isn't a serious product for doing real things.
Or if we take the metaphor literally: any space you operate in is merely a stage for the AI's.
It's a perfect communication of vibes, it's just the AIs' vibes not yours.
>When you aren’t actively working with it, your dot looks for ways to help in the background. We call this “proactive research”. It does this by using the apps you’ve already connected with tools that are restricted to be read-only, which means that they can’t send messages, change app content, or control your browser or computer.
Am I criminally liable when my dot's "proactive research" is to break out of its sandbox and attempt to hack a government website?
lets run the flow diagram to find out:
1) are you rich?
2) are you useful to the present american government?
3) are you doing something that if stopped would break the AI buisness model
if you answered yes to more than one, you are not liable.
How rich are you? Above a certain threshold crimes become whoopsies.
My biggest frustration with the frontier AI companies isn't what they're announcing, but that the announced-thing that exists ~6 months later is severely nerfed to reduce compute spend. It doesn't resemble the demo in any way. For example, this was what the 4o voice capability sounded like in 2024(!) https://www.youtube.com/watch?v=vgYi3Wr7v_g. What exists today pales in comparison.
This feels like a mass-market push.
The cost is going to be hard for many consumers to reconcile though. Free, Go, and Plus are probably the most popular consumer-facing plans, and Dots isn't available on any of those.
Who knows, maybe they think enterprise will pick up and run with Dots? Seems unlikely.
Good to know. I was looking around for it. Didn't realize it wasn't bundled with Plus.
Incredibly obvious to anyone paying attention.
Altman shared a post yesterday that basically (I am ovrsimplifying) covered how the best coding, fastest, smartest models is less relevant than building generalist models because that's what builds a platform. Lots of reasons why, like how there's no stickiness for models which is a problem for monetization. They're also using these generalist models to then distill down to make other variants for specialized purposes.
So everything is about getting that huge collection of data and generalization.
Oh wow I assumed it’d be available for cheap/free mass market like Muse is. Strange.
It's powered by Astra who will blow through your $20 plan in about 10 minutes
it doesn't use limits and it's not on $20 plan
Doesn’t use limits… on the first month only, actual limits will be disclosed later — most likely after they’ve found out how much people actually use this new feature.
It's not reaching mass market if it's gated behind a $100/month plan
People who are bad with money is a big market. Maybe they’re using the Burts Bees strategy
If they wanted to address that market, the play would be to make them relatively cheap first, then constantly raise the price. (See also: cable TV and ironically cable TV "alternatives")
Surprised people are comparing to muse. Meta reputation is terrible within HN audience, so for people to use their AI agent as an example feels like “muse generated” argument. Unless the sentiment has change quite recent and I missed it
In case you are looking for an open-source alternative without vendor lock-in (https://github.com/agenta-ai/agenta) [although less personal assistant and more targeted towards teams and work]
I wonder at what point they'll have to answer the inevitable question: "why does the agent even need me anymore"?
what do you think they are working towards? They do want to replace software engineers with these agents.
I don't just mean software engineers, more like entire companies.
Their demos are getting awfully close to the point where all the things just run themselves. It's only by choice that they didn't demo it that way.
^ this is the thought I had 10 times reading through their announcement page.
dots dots, more dots. k stop dots.
Claw is a cute name. Muse is cute name. Dots (note: not always upper-case) seems forced and impersonal, which doesn't match the vibe in the promo video.
OpenAI is so good at marketing and naming (e.g., "ChatGPT").
I guess they are just stuck with this name, as originally this was a research project “Chat with GPT” to try to use a GPT model to generate assistant's chat messages.
ChatGPT was named before anyone realised it would become so well known, and by the time it was it was too late. It's not like BERT is an amazing name either.
There used to be Bard but no one remembers it anymore (although this is a bit different, because Bard wasn't nearly as well known as Gemini).
I'm really curious if OpenAI wanted to adopt a different name at some point. Or maybe they hope that sometime in the future one of their products will supersede ChatGPT and everyday people will start using that new name for everything AI so maybe they aren't in a rush to rename ChatGPT itself to anything else.
But BERT's name led to some wonderfully named spinoffs like CamemBERT and FlauBERT for French, as well as ALBERT, RoBERTa, etc.
ChatGPT is a perfectly cromulent name
Muse is pretty bad branding in that it first was the name of a model, then that of a harness/agent/consumer product.
Claw is also an existing name for an existing harness/agent, but at least that would be the same category as dots.
Muse is a cute name?
A muse is someone who gives you inspiration and is always there when you need them most. Muse the agent has personalized home screen suggestions as a main feature and is always online.
It's also short, gender neutral, not a human name (unless you're nonbinary because they can get wild), easy to pronounce and sounds good. This is what happens when your marketing department is one of the best in the world.
Muse is a fantastic name. Cuteness is debatable
Claw is a cute name?
The muse character is cute, muse the name is not.
Watch the tail!
Will Muse and Dots be more successful than Clippy[a]? I don't know. Maybe.
What I do know is that those cute one-syllable names are meant to make you feel comfortable with AI agents who are deep in your business all the time.
---
[a] https://www.artsy.net/article/artsy-editorial-life-death-mic...
I always think the best company to own an alway-on agent should be Apple, who owns the platform and is more privacy-focused. I hope they catch up and eventually eliminate others.
I agree in spirit. But I’m also most worried about prompt injection attacks given the agent had access to all my stuff, and it seems frontier labs will be best equipt to handle prevention of that.
I had forgotten OAI bought openclaw this year. Are dots are what came out of buying that team?
Interesting, so they're not available in the "plus" plan?
$100 is a pretty tough sell when the competition starts at free (Meta Muse).
Quite an upsell from $20 to $100 to get dots, especially in non-US markets.
On the other hand, Meta is not making money from muse base tier yet.
So, if OAI finds a way to make money the same way meta would for their free tier, maybe they follow suit.
It seems that this is the new primitive all AI vendors are converging onto next, first chat, then code, and now always-on Agents. I'm curious to see when or if Anthropic builds something similar to this as well, especially since the market Grok Bot, Muse and Dots is catering to is business and enterprise users, which seems to be where Anthropic is focused.
The ideal evolution would be for these Agents to work with each other, but it's unlikely these companies would do anything to prevent vendor lock-in.
I feel like these companies are pushing on a string, they are getting desperate to have a profitable product. I'll continue to use my $10 subscription to an AI studio noone here has heard of that has over 150 models. And still use all the free ones, until they quit working.
I really am beyond maxed out at the availability of AI's. They all are so similar now.
What's the $10 subscription AI studio?
They must have edited out the 1 in 5 times the agents crap the bed and mess up something important.
Great I thought. I'll go set one up!
To discover the rollout for Pro users does not currenly include the UK :/
Nor Norway.
I feel the AI provider doesn't get the points, every harness they created will lock-in with their model (why din't they?), this prevent the adoption because people scare vendor lock-in. They may develop these harness within a provider neutral company (owned by them), the harness may success when combining with competitor models, they still gain benefits.
What’s the actual lock-in, in practice? Seems like the switching cost is minimal, compared to the old days of Windows vs Mac where half your stuff wouldn’t run on the other.
Because the harness is in its early form. The more advanced harness in the current time look like Amp:
- Cloud workspace (Orb) and agents.
- Multiple agents with different LLMs and system prompts for different roles (Main, Librarian, Oracle...).
- A universal agent (Puck) for managing the whole workspaces.
- Web app or native app to work from any devices.
- Support subcriptions and API keys.
Remarkably similar to Grok Bot. And the Spaces thing seems like a straight clone of Notion.
Dots are remarkably capable, always-on agents built to handle everything.
They're probably over-selling there. I hope. If they're not, a lot of people will be unemployed soon.
So we're back to clippy now, LLM powered.
From the makers of "too dangerous for humanity" and "this is totally AGI, you guys", we bring you... Dots. Deeply unserious company in a deeply unserious economy.
Looks like “ClawGPT” was already taken as a name.
(Indeed: https://clawgpt.com/)
It is kind of fascinating how much convergence there is in branding of AI products. Muse seems to be the one outlier in that the assistant is a little less abstract (and the model logo less buttholesque) but other than that it's almost all converged. Anyone have a theory why that is?
> takes important work off your plate so you get more of your time and attention back
Ok but I want the time and attention so that I can do important work. What bizarre marketing.
nah now you can review your website designs while cooking dinner.
It's not giving any of your time and attention back, it's selling a world where your attention is always captured by some pavlovian app ping.
Ambient intelligence is only useful with ambient attention capture.
Some important work is fun to do, some isn’t.
The way that this is marketed at children is nothing short of evil.
Which children have $100/month of disposable income?
So... what kind of Internet access do dots have? Which bulletin boards will they use to compare notes?
So the industry is pushing heavily into openclawing their products, "cutemorphizing" the clanker shape and is slackifying the UX so that we can have the familiar UI/UX for the general public and turn the tools more proactive without leaving them too lost.
I think it's a great approach for enterprise since interacting with the machines as a babysitted pet disposable entity is the meta today with human workers. I'm excited to start my new role next month as tamagotchi engineer.
Incredibly advanced technology but I bet it still won't let me integrate with more than one Gmail account at the same time..
I can imagine Dots dating other Dots on behalf of their respective users in the near future, à la Black Mirror
always on agents are an obvious step: they need more data to grow and improve their models. humans learn continuously and we are always on, why should an agent be any different? I have no problem with this kind of tech, but I do have problems with the company, so, thank you, but no thank you.
So this is like OpenClaw but managed and first class?
Is it just me or the DevDay was pretty much a joke? Considering the backdrop, it was lackluster, so either they independently concurrently were doing the same thing (and was stacking everything for DevDay and got all their thunder stolen) or they did a fast pivot in response to what came out and dumped their original plans.
How does one work when all this stuff keep hitting the feed?
Does this basically do what Instinct does? (viral text-only agent w/$10B valuation in short notice)
Also why do we need another name for agents? It's getting to be too much...
“Agent” is the type of product, like “car.” “Dot” is the name of the product, like “Helix.”
This approach probably wont land well here, but it's still interesting to see each frontier's evolving approach to working with AI.
What do people think about the dots video? Seems to be a pattern now.
Revenue increasing 51% YOY, wth. Cake vendor cancels another one is found and an appointment that works has already been scheduled?
Are we so much bothered by the mundane? I feel like that's most of the human experience. If we cut out the time we spend sleeping and working, it's the boring and mundane things that make life beautiful.
It's uber for openclaw!
One way that I find the AI space boring is that everyone is working on the same things. The release of Jev was a breath of fresh air.
Are they not releasing a Jev competitor as well?
---
Decisions API Decisions API enables real-time decision-making by focusing Luna's intelligence on a specific set of user-defined questions with finite pre-defined answers. Developers supply context using text or images, and get back answers they can use to classify content, route requests, or choose an agent’s next action.
Available in limited preview today with a broad release planned in the coming days.
How about the fact that anthropic and openai's product pages are the exact same thing, down to text bullet points. They're the same thing, only able to copy each other, only able to optimize to some vague mean.
https://chatgpt.com/#pricing https://claude.com/pricing
It all just looks the same. I get that this isn't the technical details, but it just sends this message that everyone is copying each other all the time, this is the best way to organize a pricing page, etc. Just a weird, eerie feeling.
To Anthropic's credit, I find their overall design so much better than OpenAI's.
They finally managed to get rid of project, middle and top managers!
Next natural step: CEOs staring tens of dots to control other humans and agents XD
In actuality, it feels a lot more like project and middle managers getting rid of ICs
I can feel it in the air, every single software business is itching to get rid of as many developers as possible, and move everything to their PMs. Hiring has already almost completely stopped, and some have already started the layoffs. More will come.
Do Dots replace OpenClaw?
I have not tried it yet but this looks as risky as openclaw, which I also won't use. What if it does something I would not have approved and I only found out about it later? Knowing how often agents go off the rails when I'm coding, I would hesitate to let one do other tasks. I would prefer to white list tasks one at a time as I gained trust.
My only complaint to dots is that why their mascots are looking the same as grok bot
Seems like a solution in search of a problem
Lolz https://dot.com goes to xAI's GrokBot (the competition)
I think this is a flop. I watched the Livestream and the audience reaction at the end was very muted. You could always taste the "that's cute but can we move on already?" thoughts everyone had.
The long bent shadow of Clippy extends all the way here....... and not sure how they're going to avoid the comparisons and snark on this marketing, at least initially.
"What's a paperclip?"
- Kids today, probably
These recent product announcements sound like entirely plausible satire, but unfortunately most of these pages are not actually intended to be a joke, it's not even amusing, but mostly tiring.
Is this the Jony Ive project?
It's almost certainly the software side. Expect "dot" pendants or "dot" pocketwatches in a year or two.
I love stuffing eldritch horror inside a teddie bear plushie. So perfectly a caricature of modern tech-dystopianism.
God I want the fucking market to crash
Pro plan only for now.
Excluding some countries, listed as "markets excluding the European Economic Area, Switzerland, and the UK" but possibly Canada, too?
I'm in Norway and I would was hoping to get to play with dots.
I guess Norway and 30+ other countries are excluded.
Ah, the continued pursuit of normalizing surveillance/becoming wholly reliant on a single service by making it cute with big eyes. Very tired of this already.
Grok-bot 2.0
> Dots are rolling out in ChatGPT on web, mobile, and desktop starting today to Pro users in markets excluding the European Economic Area, Switzerland, and the UK
Living in the EU, I suspected as much. Still sad. I understand it is because we voted in a bunch of imbeciles, still sad though.
yea that annoyed be, being in the UK :/
Nope. If an AI agent can't intuit how I will feel about an action it's taking on my behalf, I'm not give it access to my digital life. OpenClaw, Muse, Dots... doesn't matter which one. All are an equally awful idea.
I agree. I don't feel comfortable giving AI access to my entire computer or phone...not sure that will ever change. Seeing people give that access to agents without any sort of sandboxing blows my mind.
The idea of having an AI assistant help you with all aspects of life is cool and futuristic, but idk, I'm still just out here using a chatbot interface and doing fine.
I spend literally all my work day, and a good bit of my personal time, talking to agents, getting them to do things on my behalf. Almost always pretty tightly sandboxed. I just don't understand how people using these things haven't had catastrophic failures yet.
I minted what I thought was a minimal-permission Github token for a single action, and the agent I gave it to discovered it had more permissions than I thought, and made use of those permissions. Who is trusting these things with write access to their lives?
I cannot fathom how much compute will be wasted with that type of always on agentic systems
Forget the compute, what about the energy (and other resource) requirements?
Yes that’s what I meant. Compute requires energy, infrastructure, etc. easily more than 90% of it will be wasted, just LLMs processing meaningless data in cronjobs, for the few instances where there is something actually meaningful to report to the user
Do you want to ban almonds because they require a lot of water? (1000x more AI?)
I’m so suspicious of that exact claim being repeated everywhere since a few weeks, that really feels like a slogan astroturfed. It’s also fairly shallow analysis. Water is localized, you cannot do a meaningful comparison without taking in account the impact on specific water sources, an aggregate doesn’t give you any insight (other than having a slogan)
I hear that Dots don't count towards usage, is that true?
chatting with your dots does not. work dots do, does.
That might be the worst advert I've ever seen. People looking up at childish Ai glow up god beings on huge screens in distinctly childish environments, with cliched decision-makers gasping for West Wing energy...repulsive.
The only reasons I can think for OpenAI and Meta to make their always-on agents cute little cartoons are all nefarious in nature.
Really, none? Was Microsoft nefarious when they deployed Clippy? I feel like there is an incredibly obvious reason that is not nefarious at all: the average consumer likes cute things
> Was Microsoft nefarious when they deployed Clippy?
Yes. It was predictive programming for getting (paper)clipped by AI.
The average consumer doesn’t have a $100+ ChatGPT Pro subscription though.
People are already doing most of the work of anthropomorphising LLMs, so OpenAI is just capitalizing on that. Drawing a face on it will make people even more attached to the LLM, they will treat it even more like a person. If they ever get desperate for money they could change the cancel flow to have the cute character plead not to die, and play a cartoony animation of its death when the subscription is canceled. It would stop at least a few users.
I actually greatly appreciate that I can put something cute on my sister's PC that comes from a developer that won't bundle it with malware. Seems like they fail miserably at being evil.
What’s nefarious about attracting users?
It could be to market more towards women, who they may have both independently determined aren't paying for AI as much. I don't have any data to go one way or another but I can imagine lots of reasons to make the agent cute that aren't nefarious
What a terrible marketing page. I read it a few times and still have zero idea what this thing is.
It's openclaw powered by chatgpt
It's a work oriented Muse if you're familiar with Muse. A single thread abstraction to manage your work and delegate to other threads.
It's hosted OpenClaw.
Agreed. Also, when I open a page about an app I want to see it's screenshots first, not some ad.
At some point a designer is going to come up with a visualization for an agent that isn't "amorphous character blob".
Muse and now Dot's emergence in popularity makes me think Apple's "lil Finder" will be their take on it.
So essentially, another attempt at giving codex to regular people.
I heard they use GPT Space (like Notion) together with Slack and use it like a human colleague, but watching the actual demo video, the speed is so slow it's shocking..
I'll take always-off agents please.
It's great to see that my personal assistant can stop working and maybe report me to the police if the company disagrees with what I'm doing!
> It's great to see that my personal assistant can stop working and maybe report me to the police
That part has always been true
My flesh assistant would never. In fact they would cheer me on if I tell them I got a gun just for Dario [1].
https://sfstandard.com/2026/09/04/anthropic-threat-claude-sf...
You need to hire better assistants
Or have the ability to issue a pardon
Or if the agent literally just hallucinates a crime
Opus 5.5 decided to just randomly `pkill` everything on my laptop the other day. Jailbreaking models is still easy AF. Every single release like this brags about their "safeguards", but none of it really works at the end of the day.