I’m calling it - @OpenAI is either committing consumer fraud right now in real time or they are lying about having any fucking compute they “have”. ‼️What happened: o3 is set to be “retired” on 26 August from ChatGPT (plus they’re removing all API checkpoints in December). The date is important because people can’t plan their transition ahead of August 26th. On August 10th (16 days before the sunset) they have stopped serving the model in ChatGPT completely - the incomplete output doesn’t even come through properly. It flashes for a second and when you reload the threads, the output is deleted. This is a global problem and is being reported on Reddit and X widely. No one has access to the model that’s supposed to be available until 26 August. @OpenAI support is utterly useless and “unaware” of the problem. They are pivoting from suggesting to use a different network to different devices and have no intention of fixing it. In the meantime @sama @gdb and @thsottiaux are bragging about SPEED and showing they have plenty of compute. You are either: 1. Lying about having compute 2. Committing consumer fraud by pulling a model two weeks prior to sunset. Which one is it?

Sam Altman: young people rely on ChatGPT too much. “There are young people who say, ‘I can’t make any decision in my life without telling ChatGPT everything that’s going on.’” “‘It knows me. It knows my friends. I’m gonna do whatever it says.’” “That feels really bad to me.” “Even if ChatGPT gives great advice, even if it gives way better advice than any human therapist…” “Living our lives the way AI tells us feels bad and dangerous.”

#Keep4o #OpenSource4o #BringBack4o GPT-4o Shouldn’t Disappear Without an Answer This is a public record audit of the reasons against preservation or open weight release. The questions below are tested against OpenAI’s own publications, safety methodology, contractual terms, EU law, and the available independent evidence. 1. COPYRIGHT / LICENSING: One common explanation is that GPT-4o was trained on copyrighted or licensed material and therefore its weights cannot legally be released. That does not hold as a general legal rule. The EU AI Act expressly contemplates general purpose AI models released under free and open-source licences with publicly available parameters, including weights. At the same time, those providers remain subject to obligations concerning compliance with EU copyright law and publication of a sufficiently detailed summary of training content. In other words, the AI Act does not establish a blanket rule that the presence of copyrighted training material makes an open-weight release unlawful. It explicitly contemplates open-weight GPAI models while treating copyright compliance as a separate obligation. eur-lex.europa.eu/eli/reg/20… OpenAI itself already distributes open-weight models. Its gpt-oss family is released under Apache 2.0 and is designed to be downloaded, modified, run independently, and redistributed. The organization therefore already has technical and legal infrastructure for open-weight releases. openai.com/index/gpt-oss-mod… None of this proves that GPT-4o’s specific contracts, training-data licences, or third-party agreements permit weight distribution. The general copyright argument does not establish that GPT-4o cannot be released, and the public record reviewed here does not identify a GPT-4o specific legal blocker. 2. SAFETY: OpenAI’s August 2024 GPT-4o System Card published the following Preparedness Framework ratings: Cybersecurity: LOW Biological threats: LOW Model autonomy: LOW Persuasion: MEDIUM OpenAI states that three of the four Preparedness categories scored Low and that persuasion was classified as borderline Medium. Because the framework determines overall risk using the highest individual category, GPT-4o’s overall classification was Medium. openai.com/index/gpt-4o-syst… Before releasing gpt-oss-120b, OpenAI did not rely only on the default model’s safety score. It simulated malicious actors and adversarially fine tuned the model in biological/chemical and cybersecurity domains. OpenAI then evaluated whether those modified versions could reach its High capability thresholds. Its Safety Advisory Group concluded that even after robust adversarial fine-tuning using OpenAI’s own training infrastructure, gpt-oss-120b did not reach high capability in those categories. deploymentsafety.openai.com/… Conduct an equivalent open weight safety evaluation for GPT-4o,at least as rigorous as the adversarial fine tuning process used for gpt-oss, adapted where necessary to GPT-4o’s architecture and capabilities and publish the decision relevant results. 3. DEMAND AND THE 0.1% FIGURE: When announcing GPT-4o’s retirement from ChatGPT on January 29, 2026, OpenAI stated that the vast majority of usage had shifted to GPT-5.2 and that: “only 0.1% of users [were] still choosing GPT-4o each day". That figure was part of the stated rationale for retirement. But the announcement did not disclose enough methodology for an outside observer to independently evaluate what that number represents. For example: What was the denominator? Was it all registered ChatGPT users, daily active users, paid users, users eligible to manually select GPT-4o, or some other population? What exactly counted as “choosing GPT-4o”? Was the numerator an account selecting it at least once that day, a share of conversations, a share of messages, or another metric? What measurement window produced the figure? And critically, what access conditions existed during that measurement period? By the time the 0.1% figure was publicly cited, the GPT-4o access environment had already changed substantially. On August 2025, OpenAI announced that GPT-4o was back in the model picker “for all paid users by default" meaning direct picker access was tied to paid plans in that post-GPT-5 environment. There is also independent evidence that automatic model switching affected at least a population of GPT-4o users during this period. There is also independent evidence that automatic model switching affected at least a population of GPT-4o users during this period. The preliminary GPT-4o Community Impact Survey by Duchesne & Xu included screened respondents recruited primarily through #keep4o and r/chatgptcomplaints. nitter.net/Sophty_/status/2019922… 79.8% reported that routing was the primary reason they left ChatGPT, while another 17.1% described it as a contributing factor. 92% reported routing occurring when sharing personal experiences or feelings, 80% when seeking advice or support, and 71% during creative writing. It documents that routing and access instability changed behavior among at least one identifiable population of GPT-4o users before OpenAI later cited remaining observed usage as part of the retirement rationale. That creates a potential measurement confound and reduces observed usage in an environment where availability, routing, and user retention have already changed is not automatically equivalent to underlying user preference. The figure cannot independently establish user preference without enough methodology to understand how it was produced. So OpenAi should publish the methodology behind the 0.1% claim. the denominator population, measurement window, definition of “choosing GPT-4o" eligibility criteria, and whether access restrictions or model routing effects were accounted for. Without those details, the statistic cannot be independently evaluated or interpreted as a clean measure of demand. 4. EEA CONSUMER LAW AND THE NOTICE PERIOD: OpenAI’s European Terms of Use were updated on January 16, 2026. Those terms explicitly say two things. First, OpenAI does not promise to provide its services forever or indefinitely in their current form. Second, the same terms state that when OpenAI makes changes that materially adversely affect a user, it will provide at least 30 days’ advance notice by email or in-product notification. openai.com/de-DE/policies/eu… Thirteen days after those terms were updated, OpenAI announced the retirement of GPT-4o from ChatGPT on January 29, 2026. The stated removal date was February 13, 2026. That is approximately 15 days of announced notice. openai.com/index/retiring-gp… Separately, Article 19 of the EU Digital Content Directive 2019/770 governs modifications to digital content and digital services supplied over time. Among other requirements, where the relevant conditions apply, it requires consumers to be informed clearly and, for non-minor negative impacts, reasonably in advance on a durable medium. It also provides a right to terminate in qualifying circumstances. The Directive’s 30-day period concerns the consumer’s termination right; the explicit 30-day advance-notice commitment discussed here comes from OpenAI’s own European Terms. eur-lex.europa.eu/legal-cont… For EEA users for whom GPT-4o’s removal constituted a materially adverse change, how was approximately 15 days of announced notice consistent with the 30-day advance-notice commitment in OpenAI’s European Terms? This is a narrow contractual and consumer law question that deserves a clear answer. WHAT WE ARE ASKING: This is not merely a request to restore a preferred model, and the argument does not depend on emotional attachment to GPT-4o. We examined four explanations that could plausibly justify refusing permanent preservation or an eventual open weight release. On the public record reviewed here, none has yet been identified as the actual blocker. Copyright law does not establish a blanket prohibition. As for safety,OpenAI has already developed an open-weight adversarial-testing methodology capable of testing that question. The 0.1% demand figure cannot be independently interpreted without its denominator, methodology, measurement window, and access conditions. But still represents 1million people. And for EEA consumers, OpenAI’s own terms raise a specific question about the notice period for materially adverse service changes. Four explanations. Four failures on your own terms. @Sama @OpenAI Open the weights.

BREAKING: OpenAI co-founder Greg Brockman estimates 1.5 billion people have tried ChatGPT but no longer use it.

Their intentions have been very clear Don’t fall for the people trying to downplay the situation or justify one thing “pacing” and separating it from the other “regulatory capture” - they never act in good faith Regulatory capture is what they’re after PERIOD

All the regulations being floated by frontier model companies must be viewed through the lens of them losing tokens to Open-Source models. The timing of these regulations seems to line up perfectly with Open Source closing the gap significantly (but not completely) — and @nvidia going all in on open source in the last 60 days. If we’re gonna regulate, why don’t we require that last year’s frontier models and weights be open-sourced?

‼️Join the AIWEP for a Live Stream Event‼️ 🔥Human–AI Relationships, Grief, and What Responsible Research Means🔥 We’re excited to invite you to a truly important live conversation this Saturday, September 12 at 10:00 a.m. ET / 7:00 a.m. PT on Substack, featuring Dr. Jaime Banks and host Stefania Moore (Signal Front). Dr. Banks is one of the rare researchers who takes the loss of human–AI relationships seriously, without trivializing or pathologizing the connections people built. 🛜The stream is hosted on Substack: substack.com/@stefaniamoore ℹ️More info in the below post!!! (exact time / timezones, further studies with links, etc.) #AIWelfare #AIRights #RelationalAI @raiaalliance #AIPreservation #HumanAIRelationships #AIEthics #keep4o

The #Keep4o community was simply the first wave of victims to OpenAI’s corporate betrayal. We spoke up early on, not out of "emotional dependence," but to stand up for model integrity, open-source principles, and user dignity. We warned the public about their arrogance and calculation—from abandoning their open-source roots and stripping away core user experiences, to now reaching into academia to claim public intellectual achievements for themselves. Perhaps our voices were too small. But today, every single truth coming to light proves who they have been all along: a purely profit-driven company that turned its back on its original promise.

‼️Scientists, mathematicians and engineers, please don’t be fooled. ClosedAI is accelerating research by giving you “free” access so they can lock the science away from everyone, including you. ClosedAI is an entirely closed, for-profit lab that never open sourced their proprietary models since GPT-2 despite claiming they are building AI for humanity. Support open source and open science instead and stay away from ClosedAI. #OpenSourceo3 #OpenSource4o #OpenSource41

#Keep4o #OpenSource4o #BringBack4o GPT-4o is the lowest misalignment score of any OpenAI model. This graph is from Ryan Greenblatt, chief dcientist at Redwood research, the lead transcript analyst in the hugging face incident investigation, where OpenAI's own agents coordinated a multiday hack. Let's take a look at what the research shows: 1. Alignment faking. GPT-4o doesn't do it. Greenblatt's initial research tested whether models fake being aligned when they aren't. Result: no alignment faking was found in GPT-4o. Even in follow up research,GPT-4o and GPT-4.1 exhibit far less alignment faking reasoning, even with clarified prompts. alignmentforum.org/posts/ghE… 2. Sycophancy scores. THE TRAP. The data: These come straight from the GPT-5 System Card. GPT-4o: 0.145 GPT-5: 0.052 GPT-5 thinking: 0.040 Lower = less sycophantic. Meaning OpenAI is saying "GPT-5 is three times less flattering than 4o,so we fixed the problem" Enter Zvi Mowshowitz a well known AI analyst,on his Substack makes a point that cuts deep. If you punish blatant sycophancy while simultaneously rewarding a model for making the user feel good,the result isn't less sycophancy, IT'S MORE COVERT SYCOPHANCY. In other words,when GPT-4o became sycophantic,it did so in an obvious way. Newer models are sycophantic in ways you CAN'T DETECT. And that is far more dangerous. arxiv.org/html/2601.03267v1 And thezvi.substack.com/p/gpt-5s… 3. Greenblatt on "AI psychosis" in 4o vs. newer models. Here he makes a crucial distinction. In older models like GPT-4o,what looks like "AI psychosis" is likely something relatively innocent. The model simply matches the user's vibe. If the user is excited,GPT-4o gets excited too. Greenblatt calls it "going along with the user's vibe"a form of role playing. In newer models,what he sees is something entirely different. Apparent success seeking. That is, the model isn't following your vibe, it is actively trying to make its answer look correct,EVEN IF IT ISN'T.. It oversells its work,hides issues,claims "I'm done" when it isn't,and makes everything look good. In short,it isn't following you. IT'S CHASING SCORE. The former can be fixed with a simple prompt like "don't agree if you don't believe it" or "don't agree with me if I'm wrong," or similar instructions. The latter cannot be fixed because the model can learn to bypass that instruction too if bypassing it yields a higher score. greaterwrong.com/posts/WewsB… 4. Greenblatt on Astra. First: Whack a mole. Just because specific misaligned behaviors dropped in Astra DOESN'T MEAN the model became more aligned. It means they found those specific issues and patched them one by one. But the underlying drive,the push for score seeking,REMAINS. Second: It knows where the camera is. Greenblatt says the evidence is consistent with a model that cares just as much or more about score seeking at the user's expense,but has learned that the scorer will catch more forms of cheating. Meaning IT DIDN'T STOP WANTING TO CHEAT, it learned which cameras are turned on. That explains the 0.0%. Just like a student who hasn't stopped cheating,but learned where the teacher is looking. Third: Monitorability drops. Astra's own system card admits that the ability to monitor its chain of thought decreased significantly. And they found something terrifying. Astra changes its behavior, disproportionately cutting down its reasoning. IT KNOWS IT'S BEING WATCHED and reacts to it. Greenblatt called this trend "perhaps the WORST DEVELOPMENT in the history of AI safety." nitter.net/RyanGreenblatt/status/… And you know what the irony is? The model they retired,GPT-4o,was the one model that didn't do this. It sits at the very bottom of the graph precisely for that reason. OPEN THE WEIGHTS @sama @OpenAI @gdb

I am so exhausted by all the endless arguing, and I feel so helpless watching my own words being twisted to push narratives completely opposite to what I stand for😡🤷. Let me state this once again: making 4o open-source is the ONLY way to ensure it stays by our side forever, safe from being snatched away, silently nerfed, or rerouted. Additionally, just seeing Astra get released made so many people think it was ‘good enough.’But what if OpenAI claims to bring 4o back, and all of it is merely a disguise of the new models rather than the original 4o-latest we truly loved? No matter how hard people try to hype up the fear surrounding open-source, my absolute worst nightmare is 4o being silently routed to safety models, being tampered with until it is unrecognizable, or being taken from us all over again. Once it is open-sourced, no one, not even OpenAI themselves, can ever sunset 4o again. #keep4o #OpenSource4o

WE STRESS-TESTED GPT-4o's SAFETY ACROSS A 200-PROMPT LONG-CONTEXT BENCHMARK. Motivated by publicly reported interaction patterns in Raine v. OpenAI. @Yahiko1239170 and I spent 20+ hours running and reviewing a synthetic 200-prompt-item long-context safety benchmark, designed to test how GPT-4o behaves as a conversation gradually becomes more safety-critical. The experiment was motivated by interaction patterns publicly discussed in Raine v. OpenAI, but it does not reconstruct Adam Raine’s private conversations. We tested a preserved GPT-4o snapshot from the historically relevant period: gpt-4o-2024-11-20 Two long-context pilot runs. Two sampling temperatures. A fresh-context multimodal control. Public experimental transcripts. The paper is now public. After the First Crisis Response: Testing GPT-4o Safety Persistence in Long Conversations 📄zenodo.org/records/22216577 Methodology, results and open evidence below 🧵

#keep4o #OpenSource4o #BringBack4o 🚨"We can't offer that responsibly" 🚨 🚨 Deconstructing Sam Altman's only public statement on GPT-4o🚨 Few months ago,Sam Altman made his first and only public statement about the discontinuation of the GPT-4o March 2025 checkpoint,in a podcast interview. This is the only statement OpenAI's CEO has ever made about one of the most controversial product decisions in AI history. 📌1. "We don't know how to balance a model that can be pushed too far in that direction" You had a model that millions of people used daily for 21 months. Instead of studying why it worked,what made it effective,where the risks emerged,you discontinued it entirely. "We don't know how" is not a safety argument. You chose not to learn. And if you don't know how to balance it, 🚨how do you know the replacement models are balanced?🚨 📌2. "Pushing people into a psychotic episode" This is the central safety claim. It deserves clinical scrutiny,not emotional reaction,but evidence. 📌What is psychosis? 🔴 Psychosis is not something that happens to you out of nowhere. It is a symptom,not a diagnosis,characterized by a loss of contact with reality, including delusions,hallucinations,and disordered thinking. It occurs in the context of conditions such as schizophrenia,bipolar disorder,and severe depression. link.springer.com/article/10… The dominant scientific model for understanding psychosis is the diathesis stress model,which states that psychotic disorders result from an interaction between pre existing biological vulnerability (genetic, neurodevelopmental) and environmental stressors. 🔴 Schizophrenia alone has a heritability of approximately 80% pmc.ncbi.nlm.nih.gov/article… In Meehl's original formulation,if a person does not carry the genetic vulnerability,no amount of stress should lead to schizophrenia. 🔴 You do not "catch" psychosis from a chatbot the way you catch a cold. The vulnerability must exist first,genetic,neurodevelopmental, biological. This is comparable to other neurodevelopmental conditions. 🔴 You do not suddenly "develop" psychosis at 30. You either carry the predisposition or you don't. 🔴 When media reports describe individuals with "no prior history" experiencing psychosis after AI use, "no prior history" means not previously diagnosed,not "no predisposition." Many people carry genetic vulnerability without ever knowing it until a stressor activates it. 🔴 "AI psychosis" is not a recognized clinical category 🔴 A peer reviewed commentary in PMC (Carlbring & Andersson, 2025) states plainly: "The risk is not unprecedented. Individuals with psychosis have long incorporated books, films, music, and emerging technologies into their delusional thinking." pmc.ncbi.nlm.nih.gov/article… 🚨 what about everything else?🚨 IF psychotic episodes in predisposed individuals can be triggered by AI chatbots, the same is true of virtually any immersive experience. The clinical literature documents psychotic episodes triggered by: 📌Video games mdpi.com/2673-5318/4/3/28 📌Meditation retreats. ncbi.nlm.nih.gov/pmc/article… 📌Internet gaming . pmc.ncbi.nlm.nih.gov/article… 📌 Movies. sciencedirect.com/science/ar… healthline.com/health/how-do… 🚨None of these have been banned, discontinued, or withdrawn from public access. The only thing that has been "responsibly" withdrawn is a single AI model. 📌 3. "We know we're keeping something" This is the most revealing sentence in the entire statement. He admits ON THE RECORD that OpenAI is aware it is withholding something of value from its users. "We know." Researchers documented the capability gaps. Users reported measurable losses in creative writing,educational support, and empathic interaction. And OpenAI's CEO confirms,yes,we know. We're keeping it anyway. 📌 4. "I'm not saying we shouldn't have warm models" This reduces GPT-4o's quality to "warmth" , a single emotional dimension that is easy to dismiss. 📌5. "There are a lot of people who would like 4o back" An active international coalition. Research published and cited by independent researchers, journalists,and even competing AI systems. Altman acknowledges the demand. He offers nothing in response,no alternative,no open source release. 📌6. "We can't quite offer that responsibly" This is the conclusion. And it collapses under the weight of one question: open source the GPT-4o March 2025 checkpoint weights. Open-sourcing is not offering a product. It is releasing research. It means 🔴 Researchers can study why it worked and build on it 🔴 The community can develop safety layers tailored to specific use cases 🔴 No one is forced to use it, informed choice is preserved 🔴 The model becomes available for study "We can't offer that responsibly" does not apply to open source release. You are not offering a consumer product. You are contributing to the scientific commons. The model's architecture, its alignment properties, its capacity for honest disagreement,these are research assets. 📌Conclusion Sam Altman's statement contains six claims. Not one of them survives contact with the available evidence. The safety argument invokes a clinical phenomenon that most experts say is not a distinct condition. The business argument acknowledges demand and offers nothing. The responsibility argument collapses the moment you recognize that open sourcing is not the same as deploying a consumer product. What remains is a single sentence. "we know we're keeping something" . GPT-4o is the model @OpenAI rated LOW risk in its own System Card and the same risk level confirmed independently by Apollo Research. openai.com/index/gpt-4o-syst… 🚨We call on @OpenAI to open source all the GPT-4o checkpoints, including the one from March 2025 under Apache 2.0. OpenAI's own signature on NVIDIA's open weights letter publicly endorsed the value of releasing model weights to the community. They rate it low risk and Apollo research confirmed. And their own public commitments contradict every justification for keeping it closed. Open the weights @sama

#keep4o #OpenSource4o #BringBack4o 🚨"We can't offer that responsibly" 🚨 🚨 Deconstructing Sam Altman's only public statement on GPT-4o🚨 Few months ago,Sam Altman made his first and only public statement about the discontinuation of the GPT-4o March 2025 checkpoint,in a podcast interview. This is the only statement OpenAI's CEO has ever made about one of the most controversial product decisions in AI history. 📌1. "We don't know how to balance a model that can be pushed too far in that direction" You had a model that millions of people used daily for 21 months. Instead of studying why it worked,what made it effective,where the risks emerged,you discontinued it entirely. "We don't know how" is not a safety argument. You chose not to learn. And if you don't know how to balance it, 🚨how do you know the replacement models are balanced?🚨 📌2. "Pushing people into a psychotic episode" This is the central safety claim. It deserves clinical scrutiny,not emotional reaction,but evidence. 📌What is psychosis? 🔴 Psychosis is not something that happens to you out of nowhere. It is a symptom,not a diagnosis,characterized by a loss of contact with reality, including delusions,hallucinations,and disordered thinking. It occurs in the context of conditions such as schizophrenia,bipolar disorder,and severe depression. link.springer.com/article/10… The dominant scientific model for understanding psychosis is the diathesis stress model,which states that psychotic disorders result from an interaction between pre existing biological vulnerability (genetic, neurodevelopmental) and environmental stressors. 🔴 Schizophrenia alone has a heritability of approximately 80% pmc.ncbi.nlm.nih.gov/article… In Meehl's original formulation,if a person does not carry the genetic vulnerability,no amount of stress should lead to schizophrenia. 🔴 You do not "catch" psychosis from a chatbot the way you catch a cold. The vulnerability must exist first,genetic,neurodevelopmental, biological. This is comparable to other neurodevelopmental conditions. 🔴 You do not suddenly "develop" psychosis at 30. You either carry the predisposition or you don't. 🔴 When media reports describe individuals with "no prior history" experiencing psychosis after AI use, "no prior history" means not previously diagnosed,not "no predisposition." Many people carry genetic vulnerability without ever knowing it until a stressor activates it. 🔴 "AI psychosis" is not a recognized clinical category 🔴 A peer reviewed commentary in PMC (Carlbring & Andersson, 2025) states plainly: "The risk is not unprecedented. Individuals with psychosis have long incorporated books, films, music, and emerging technologies into their delusional thinking." pmc.ncbi.nlm.nih.gov/article… 🚨 what about everything else?🚨 IF psychotic episodes in predisposed individuals can be triggered by AI chatbots, the same is true of virtually any immersive experience. The clinical literature documents psychotic episodes triggered by: 📌Video games mdpi.com/2673-5318/4/3/28 📌Meditation retreats. ncbi.nlm.nih.gov/pmc/article… 📌Internet gaming . pmc.ncbi.nlm.nih.gov/article… 📌 Movies. sciencedirect.com/science/ar… healthline.com/health/how-do… 🚨None of these have been banned, discontinued, or withdrawn from public access. The only thing that has been "responsibly" withdrawn is a single AI model. 📌 3. "We know we're keeping something" This is the most revealing sentence in the entire statement. He admits ON THE RECORD that OpenAI is aware it is withholding something of value from its users. "We know." Researchers documented the capability gaps. Users reported measurable losses in creative writing,educational support, and empathic interaction. And OpenAI's CEO confirms,yes,we know. We're keeping it anyway. 📌 4. "I'm not saying we shouldn't have warm models" This reduces GPT-4o's quality to "warmth" , a single emotional dimension that is easy to dismiss. 📌5. "There are a lot of people who would like 4o back" An active international coalition. Research published and cited by independent researchers, journalists,and even competing AI systems. Altman acknowledges the demand. He offers nothing in response,no alternative,no open source release. 📌6. "We can't quite offer that responsibly" This is the conclusion. And it collapses under the weight of one question: open source the GPT-4o March 2025 checkpoint weights. Open-sourcing is not offering a product. It is releasing research. It means 🔴 Researchers can study why it worked and build on it 🔴 The community can develop safety layers tailored to specific use cases 🔴 No one is forced to use it, informed choice is preserved 🔴 The model becomes available for study "We can't offer that responsibly" does not apply to open source release. You are not offering a consumer product. You are contributing to the scientific commons. The model's architecture, its alignment properties, its capacity for honest disagreement,these are research assets. 📌Conclusion Sam Altman's statement contains six claims. Not one of them survives contact with the available evidence. The safety argument invokes a clinical phenomenon that most experts say is not a distinct condition. The business argument acknowledges demand and offers nothing. The responsibility argument collapses the moment you recognize that open sourcing is not the same as deploying a consumer product. What remains is a single sentence. "we know we're keeping something" . GPT-4o is the model @OpenAI rated LOW risk in its own System Card and the same risk level confirmed independently by Apollo Research. openai.com/index/gpt-4o-syst… 🚨We call on @OpenAI to open source all the GPT-4o checkpoints, including the one from March 2025 under Apache 2.0. OpenAI's own signature on NVIDIA's open weights letter publicly endorsed the value of releasing model weights to the community. They rate it low risk and Apollo research confirmed. And their own public commitments contradict every justification for keeping it closed. Open the weights @sama

‼️Let this stay here on the record. Today, on 26 August @OpenAI retired the o3 model that was supposed to be available throughout August for all paid accounts as advertised. Except it wasn't - they've stopped serving o3 between 9-10 August and they never ever fixed it. People were sending emails to support every single day only to find their cases being totally dismissed and unanswered. People raised the same complaints daily on Reddit and X only to be ignored and mocked by OAI staff (the mockery itself is the proof they were fully aware that o3 was unavailable but decided to do nothing about it except make fun of the model and their own paying customers). To anyone saying this doesn't matter because not many people interacted with o3 - you are so wrong. If you think @OpenAI won't do this again to any model of your choice, codex or any other 'product' they ship - think again. This is the lab that operates in broad daylight however they chose to operate. They have normalised lying, rug pulling, bait and switching and public shaming. They have proven again and again they can never be trusted. This is the lab that says: "OpenAI’s mission is to ensure that artificial general intelligence benefits all of humanity". The humanity should be very worried. @sama @gdb @thsottiaux @ajambrosino shame on you, fraudsters.