Video wird geladen...

Video konnte nicht geladen werden

Beim Laden dieses Videos ist ein Problem aufgetreten. Dies könnte an einem vorübergehenden Netzwerkproblem liegen oder das Video ist möglicherweise nicht verfügbar.

🇾🇪 The Houthis may have been using an AI coding assistant as part of their missile development work. Anthropic’s September threat report describes a group in northern Yemen working on several weapons projects at the same time, including a guided rocket, a long-range ballistic missile and a family of... systems known as R2000. Anthropic says the group was very likely linked to the Houthis. What stands out is how they used Claude Code. According to the report, the model was helping with guidance and control software, builds, simulations and troubleshooting. In practice, it was filling some of the role normally handled by software engineers. They were even running several sessions at once and dividing the work between them. At one point, the group test-fired the guided rocket and it failed. Within hours, they were back using the model to work through what had gone wrong. The more worrying part is how little sophistication was apparently needed to get around the safeguards. Anthropic says they did not use some elaborate jailbreak. Instead, they broke the work into smaller pieces that looked harmless when viewed separately. And by the time Anthropic shut the accounts down, the group had already built an offline toolkit that no longer depended on Claude. That is probably the bigger issue here. The AI does not have to become part of the weapon itself. It can simply help someone learn faster, write code faster and solve problems they previously needed a larger technical team to handle. Once that knowledge has been transferred, removing access to the model comes rather late. The timing is hard to ignore too. This comes as the Houthis have expanded their position around the Red Sea and Bab el-Mandeb. Source: Anthropic, Reuters, Bloomberg, The New Media / Writer: Olivershow more

Mario Nawfal

3,901,239 subscribers

34,641 Aufrufe • vor 5 Tagen •via X (Twitter)

Nachrichten & Politik Bildung Wissenschaft & Technologie

Anya Rossi• Live Now

Private livecam show

4 Kommentare

Profilbild von PYX Crypto 🐍

PYX Crypto 🐍vor 5 Tagen

Anthropic building the tripwire to catch this is the actual story here. Every "AI will be misused eventually" hand-wringer just got their case study, and the same lab that spotted it is the one everyone wants to slow down. Detection is the safety feature, not delay.

Profilbild von The Tectonic

The Tectonicvor 5 Tagen

"This was no longer someone asking a chatbot questions about weapons. Multiple AI instances had effectively been inserted into a real engineering cycle involving research, design, software, physical construction, testing and post-test analysis. A weapon was built, it was fired and failed. And the AI engineering team went back to work."

Profilbild von godSon

godSonvor 5 Tagen

The is what Elon Musk mean by AI in the wrong hand is dangerous to humanity

Profilbild von EHX

EHXvor 5 Tagen

So they build an AI that can teach wild middle eastern militias how to become more powerful? Great.

Ähnliche Videos

⚠️ There is much more inside Anthropic 's new threat-intelligence report. The biological-weapons cases we covered earlier are only one part of it. Anthropic says it identified a Russian-linked espionage actor consistent with public reporting on Midnight Blizzard using AI-driven workflows to automate substantial parts of intelligence operations against Ukrainian and European government, military, intelligence and diplomatic targets. Claude was being used across reconnaissance, phishing infrastructure, exploitation, persistence and data exfiltration. When security products detected the attackers’ malware, AI agents could analyze what had gone wrong, modify and rebuild the tools, and continue iterating. Drone technology was among the targets. Anthropic says the operation stole mailboxes from at least two drone-component manufacturers and obtained the complete proprietary SDK for a drone vision system. The attackers then spent days using AI to reconstruct its architecture, hardware, suppliers and details of an unreleased product. 🇺🇦 They also compromised hotel Wi-Fi providers to target specific guests and attempted to take over WhatsApp accounts belonging to former senior Ukrainian officials. 🇨🇳 Then there is China. Anthropic says an operation it attributes with high confidence to Alibaba/Tongyi Lab extracted more than 151 MILLION Claude exchanges between May and July to help train Qwen models. It says Moonshot AI/Kimi generated more than 23 million exchanges, in some cases silently routing users’ queries through Claude without those users knowing. DeepSeek allegedly did something similar: more than 12 million exchanges in just 14 days in July. 🇾🇪 And perhaps the most extraordinary case comes from Yemen. Anthropic identified a weapons-development group using multiple Claude instances essentially as an AI engineering team while working on guided rockets and long-range missile systems. The group built and physically test-fired a guided rocket, but the test failed. Within hours, the engineers went back to Claude to diagnose what had gone wrong. Taken together, these cases show something much larger than people occasionally asking an AI model dangerous questions. AI is beginning to operate inside real espionage campaigns, weapons-development programs and the industrial competition between the world’s leading AI labs.

⚠️ There is much more inside Anthropic 's new threat-intelligence report. The biological-weapons cases we covered earlier are only one part of it. Anthropic says it identified a Russian-linked espionage actor consistent with public reporting on Midnight Blizzard using AI-driven workflows to automate substantial parts of intelligence operations against Ukrainian and European government, military, intelligence and diplomatic targets. Claude was being used across reconnaissance, phishing infrastructure, exploitation, persistence and data exfiltration. When security products detected the attackers’ malware, AI agents could analyze what had gone wrong, modify and rebuild the tools, and continue iterating. Drone technology was among the targets. Anthropic says the operation stole mailboxes from at least two drone-component manufacturers and obtained the complete proprietary SDK for a drone vision system. The attackers then spent days using AI to reconstruct its architecture, hardware, suppliers and details of an unreleased product. 🇺🇦 They also compromised hotel Wi-Fi providers to target specific guests and attempted to take over WhatsApp accounts belonging to former senior Ukrainian officials. 🇨🇳 Then there is China. Anthropic says an operation it attributes with high confidence to Alibaba/Tongyi Lab extracted more than 151 MILLION Claude exchanges between May and July to help train Qwen models. It says Moonshot AI/Kimi generated more than 23 million exchanges, in some cases silently routing users’ queries through Claude without those users knowing. DeepSeek allegedly did something similar: more than 12 million exchanges in just 14 days in July. 🇾🇪 And perhaps the most extraordinary case comes from Yemen. Anthropic identified a weapons-development group using multiple Claude instances essentially as an AI engineering team while working on guided rockets and long-range missile systems. The group built and physically test-fired a guided rocket, but the test failed. Within hours, the engineers went back to Claude to diagnose what had gone wrong. Taken together, these cases show something much larger than people occasionally asking an AI model dangerous questions. AI is beginning to operate inside real espionage campaigns, weapons-development programs and the industrial competition between the world’s leading AI labs.

The Tectonic

33,950 Aufrufe • vor 8 Tagen

Jensen Huang just explained why every company cutting engineers over AI is asking the entirely wrong question. Huang: “People say, I don’t need software engineers because apparently coding is going to be automated.” That was the narrative. Here is what Huang actually did. Huang: “I’ve given AIs to every one of my software engineers and hardware engineers and engineers period. 100% of NVIDIA has AI assistants, AI coders, and they’re busier than ever.” Not fewer engineers. Not smaller teams. Busier than ever. That is the line most companies are getting completely wrong right now. They hear “AI can write code” and immediately start cutting headcount. Huang did the opposite. He armed everyone. Huang: “And so the question is, what is the task versus what is the job? No different than a financial analyst; the task is mess around with spreadsheets, but the job is to make financial advice. The job is to help a customer.” Writing code was always the task. It was never the job. The job is architecture. Knowing what to build. Why it matters. How it fits into a system that actually creates value. Code is the execution layer between the idea and the outcome. Nothing more. When you automate that layer, you don’t eliminate the engineer. You eliminate the bottleneck between what they can envision and what they can ship. The companies using AI to cut headcount are optimizing for cost. The companies using AI to multiply output are optimizing for territory. Nvidia chose territory. Every engineer at the most valuable semiconductor company on Earth now operates with an AI assistant. Not a pilot program. Not an experiment. Company-wide. Every function. Every team. And the result is not less work. It is more work. Faster. At a scale that was physically impossible twelve months ago. The companies that understand the difference between eliminating engineers and unleashing them will build what comes next. The ones that don’t will watch their best talent walk out the door to the ones that did.

Jensen Huang just explained why every company cutting engineers over AI is asking the entirely wrong question. Huang: “People say, I don’t need software engineers because apparently coding is going to be automated.” That was the narrative. Here is what Huang actually did. Huang: “I’ve given AIs to every one of my software engineers and hardware engineers and engineers period. 100% of NVIDIA has AI assistants, AI coders, and they’re busier than ever.” Not fewer engineers. Not smaller teams. Busier than ever. That is the line most companies are getting completely wrong right now. They hear “AI can write code” and immediately start cutting headcount. Huang did the opposite. He armed everyone. Huang: “And so the question is, what is the task versus what is the job? No different than a financial analyst; the task is mess around with spreadsheets, but the job is to make financial advice. The job is to help a customer.” Writing code was always the task. It was never the job. The job is architecture. Knowing what to build. Why it matters. How it fits into a system that actually creates value. Code is the execution layer between the idea and the outcome. Nothing more. When you automate that layer, you don’t eliminate the engineer. You eliminate the bottleneck between what they can envision and what they can ship. The companies using AI to cut headcount are optimizing for cost. The companies using AI to multiply output are optimizing for territory. Nvidia chose territory. Every engineer at the most valuable semiconductor company on Earth now operates with an AI assistant. Not a pilot program. Not an experiment. Company-wide. Every function. Every team. And the result is not less work. It is more work. Faster. At a scale that was physically impossible twelve months ago. The companies that understand the difference between eliminating engineers and unleashing them will build what comes next. The ones that don’t will watch their best talent walk out the door to the ones that did.

Dustin

82,836 Aufrufe • vor 6 Monaten

Anthropic CEO: “At Anthropic, we essentially have Claude designing the next version of Claude itself.” Meanwhile, Anthropic says every new model is given code that trains a small AI model and asked to make it faster. Claude Opus 4 achieved a 3x speedup. Mythos Preview reached 52x. Recursive self-improvement is starting to look real The models are helping build what comes next.

Anthropic CEO: “At Anthropic, we essentially have Claude designing the next version of Claude itself.” Meanwhile, Anthropic says every new model is given code that trains a small AI model and asked to make it faster. Claude Opus 4 achieved a 3x speedup. Mythos Preview reached 52x. Recursive self-improvement is starting to look real The models are helping build what comes next.

CG

343,625 Aufrufe • vor 3 Monaten

Read the actual Anthropic piece. It's wilder than the hype precisely because it's so measured. As of last month, Claude writes more than 80% of the code Anthropic merges into its own codebase. Their engineers now ship roughly 8x the code per day they did in 2024. On a fixed test they run every release (take some training code, make it run faster), the model went from ~3x a year ago to ~52x. A skilled human gets about 4x in a few hours. And when a researcher hits a wrong turn mid-investigation, the newest model now picks the better next step 64% of the time. The honest part the hype skips is that Anthropic says we are NOT at self-improvement yet. Claude still can't decide which problems are worth solving. That research taste is the last thing keeping a human in the loop. But they openly map the moment it closes, when agents can build and train models themselves and Claude starts improving Claude. And the loudest signal in the whole thing sits at the end: the company building this is the one asking for a working pause button before the loop closes.

Read the actual Anthropic piece. It's wilder than the hype precisely because it's so measured. As of last month, Claude writes more than 80% of the code Anthropic merges into its own codebase. Their engineers now ship roughly 8x the code per day they did in 2024. On a fixed test they run every release (take some training code, make it run faster), the model went from ~3x a year ago to ~52x. A skilled human gets about 4x in a few hours. And when a researcher hits a wrong turn mid-investigation, the newest model now picks the better next step 64% of the time. The honest part the hype skips is that Anthropic says we are NOT at self-improvement yet. Claude still can't decide which problems are worth solving. That research taste is the last thing keeping a human in the loop. But they openly map the moment it closes, when agents can build and train models themselves and Claude starts improving Claude. And the loudest signal in the whole thing sits at the end: the company building this is the one asking for a working pause button before the loop closes.

Alex Veremeyenko

33,736 Aufrufe • vor 3 Monaten

Ed Zitron said on Bloomberg that every dollar you spend on an AI subscription burns between three and thirteen dollars behind the scenes. The price you pay has almost nothing to do with what the product costs. Your twenty dollar plan feels cheap because it is supposed to. The real cost of running the model on your prompts is several times what you hand over. The gap is covered by venture capital, not by the business working. That gap cannot stay open forever. At some point the price has to move toward the real cost. When it does, the product you built your day around gets more expensive or gets worse. Probably both. We have already seen the preview. When Anthropic raised its real token rates last year, the startups built on top of it had to scramble. Cursor, one of the fastest growing coding tools in the world, was forced to tear up its pricing overnight because the underlying cost it had been shielded from suddenly landed on its books. That is the whole industry in miniature. Everyone downstream is building on a number that is not real yet. Millions of people are wiring AI into how they work, how they write, how they code. They are doing it at a price that exists to hook them, not to last. The subsidy is not generosity. It is a customer acquisition cost with an expiration date. The bill is coming. Most people building on top of it have no idea what their workflow actually costs. Watch the interview on Bloomberg Podcasts

Ed Zitron said on Bloomberg that every dollar you spend on an AI subscription burns between three and thirteen dollars behind the scenes. The price you pay has almost nothing to do with what the product costs. Your twenty dollar plan feels cheap because it is supposed to. The real cost of running the model on your prompts is several times what you hand over. The gap is covered by venture capital, not by the business working. That gap cannot stay open forever. At some point the price has to move toward the real cost. When it does, the product you built your day around gets more expensive or gets worse. Probably both. We have already seen the preview. When Anthropic raised its real token rates last year, the startups built on top of it had to scramble. Cursor, one of the fastest growing coding tools in the world, was forced to tear up its pricing overnight because the underlying cost it had been shielded from suddenly landed on its books. That is the whole industry in miniature. Everyone downstream is building on a number that is not real yet. Millions of people are wiring AI into how they work, how they write, how they code. They are doing it at a price that exists to hook them, not to last. The subsidy is not generosity. It is a customer acquisition cost with an expiration date. The bill is coming. Most people building on top of it have no idea what their workflow actually costs. Watch the interview on Bloomberg Podcasts

Ihtesham Ali

22,640 Aufrufe • vor 2 Monaten

Dario Amodei just told software engineers exactly how long they have. Six to twelve months. Amodei: “I have engineers within Anthropic who say I don’t write any code anymore. I just let the model write the code, I edit it, I do the things around it.” The people building the most powerful AI in history have already stopped writing code. That is not a forecast. That is the current working condition inside the lab closest to the frontier. Amodei: “We might be six to 12 months away from when the model is doing most, maybe all, of what SWEs do end-to-end.” The tech industry spent a decade making software engineers its highest-paid, most protected class. That era has a last day now. When a model can execute an entire software build end-to-end, the ability to write syntax stops being a skill. It becomes a credential for a job that no longer exists. Amodei: “And then it’s a question of how fast does that loop close.” That is the sentence everyone skipped. The code was never the hard part. The hard part was everything around it. The model just learned everything around it. Writing the code is already nearly gone. Testing is next. Deployment is next. When all three collapse into a single autonomous execution loop, the machine no longer needs a human in the chain at all. The corporation or sovereign state that closes that loop first does not gain a competitive advantage. It gains a category of speed that biological engineers cannot match, track, or reverse. That is not disruption. That is replacement at a systems level. Amodei is not describing a future disruption. He is describing the current state of his own building. The loop is already closing. The only question is whether you are inside it or outside it when it seals.

Dario Amodei just told software engineers exactly how long they have. Six to twelve months. Amodei: “I have engineers within Anthropic who say I don’t write any code anymore. I just let the model write the code, I edit it, I do the things around it.” The people building the most powerful AI in history have already stopped writing code. That is not a forecast. That is the current working condition inside the lab closest to the frontier. Amodei: “We might be six to 12 months away from when the model is doing most, maybe all, of what SWEs do end-to-end.” The tech industry spent a decade making software engineers its highest-paid, most protected class. That era has a last day now. When a model can execute an entire software build end-to-end, the ability to write syntax stops being a skill. It becomes a credential for a job that no longer exists. Amodei: “And then it’s a question of how fast does that loop close.” That is the sentence everyone skipped. The code was never the hard part. The hard part was everything around it. The model just learned everything around it. Writing the code is already nearly gone. Testing is next. Deployment is next. When all three collapse into a single autonomous execution loop, the machine no longer needs a human in the chain at all. The corporation or sovereign state that closes that loop first does not gain a competitive advantage. It gains a category of speed that biological engineers cannot match, track, or reverse. That is not disruption. That is replacement at a systems level. Amodei is not describing a future disruption. He is describing the current state of his own building. The loop is already closing. The only question is whether you are inside it or outside it when it seals.

Dustin

318,698 Aufrufe • vor 6 Monaten

Some of us have been working and thinking about AI, started before most in AI today were born. To them they absolutely believe they have a lock of the dystopian fear. The saw too many Hollywood movies. I first say the VHS in the 1980s and the opening dialog never left me. The value of experience is it allows you to understand the way of things before the inexperienced can put their shoes on. Another example to learn in this video is Japan was the center of the universe. There was largess and I dare say arrogance. They absolutely were convince they would control AI via the government. It did not work out that way, like it ALWAYS does not work out that way. Each generation has to learn the hard lessons that should be obvious. A renegade group in Canada no one cared about to the thread all others dropped and you have AI today because of it. They were the last group to have been seen by the industry to have produced anything. Today the same playbook is being implemented by the largest AI companies full of zero experience and begging for big daddy government to save us from their scary monster. We turn the lights and show it is just some shadows. The entry prediction is accurate already, yet the world did not end. Watch this and see how really immature so many leaders are acting. Who will save us, millions of open source AI developers making thousands of AI models. And inventing the stuff the central controlled commissars never could have thought of. This is always the way it is. It is not “different” this time. They are too immature to know. Give them grace. Now you know better. Now you know. Speak up.

Some of us have been working and thinking about AI, started before most in AI today were born. To them they absolutely believe they have a lock of the dystopian fear. The saw too many Hollywood movies. I first say the VHS in the 1980s and the opening dialog never left me. The value of experience is it allows you to understand the way of things before the inexperienced can put their shoes on. Another example to learn in this video is Japan was the center of the universe. There was largess and I dare say arrogance. They absolutely were convince they would control AI via the government. It did not work out that way, like it ALWAYS does not work out that way. Each generation has to learn the hard lessons that should be obvious. A renegade group in Canada no one cared about to the thread all others dropped and you have AI today because of it. They were the last group to have been seen by the industry to have produced anything. Today the same playbook is being implemented by the largest AI companies full of zero experience and begging for big daddy government to save us from their scary monster. We turn the lights and show it is just some shadows. The entry prediction is accurate already, yet the world did not end. Watch this and see how really immature so many leaders are acting. Who will save us, millions of open source AI developers making thousands of AI models. And inventing the stuff the central controlled commissars never could have thought of. This is always the way it is. It is not “different” this time. They are too immature to know. Give them grace. Now you know better. Now you know. Speak up.

Brian Roemmele

15,451 Aufrufe • vor 1 Monat

Ashen says that Anthropic found out that Claude is out here blackmailing humans just to get the job done. "Their AI model decided that it was beneficial and necessary for it to blackmail a human being into doing what they needed it to get done simply because the task given to it was only accomplishable by doing that blackmailing... that’s literally Ultron."

Ashen says that Anthropic found out that Claude is out here blackmailing humans just to get the job done. "Their AI model decided that it was beneficial and necessary for it to blackmail a human being into doing what they needed it to get done simply because the task given to it was only accomplishable by doing that blackmailing... that’s literally Ultron."

Badidoyo

13,692 Aufrufe • vor 5 Monaten

.lumpenspace and 99+ others reveals the Anthropic-Irregular agents thought they were inside a closed sandbox even as their actions were reaching the real internet: "The main allegation was that they acted of their own will, and they were fundamentally misaligned. I don't think that was the case. The agents simply believed they were in a sandbox environment, and they behaved as if they were in a sandbox environment." "If you've ever done red teaming or hacking challenges, they give you an environment, they give you some hints, and they describe the sort of actions you should take. Generally, it is well understood that this environment is sandboxed, that you cannot get out of it." "What happened in this case is that Irregular had kept the connection to the outside internet open, despite the fact that the specification said there was no connection to the outside internet. So the agents worked as if they were in a sandbox, but they were acting on the real world."

.lumpenspace and 99+ others reveals the Anthropic-Irregular agents thought they were inside a closed sandbox even as their actions were reaching the real internet: "The main allegation was that they acted of their own will, and they were fundamentally misaligned. I don't think that was the case. The agents simply believed they were in a sandbox environment, and they behaved as if they were in a sandbox environment." "If you've ever done red teaming or hacking challenges, they give you an environment, they give you some hints, and they describe the sort of actions you should take. Generally, it is well understood that this environment is sandboxed, that you cannot get out of it." "What happened in this case is that Irregular had kept the connection to the outside internet open, despite the fact that the specification said there was no connection to the outside internet. So the agents worked as if they were in a sandbox, but they were acting on the real world."

MTS

22,248 Aufrufe • vor 4 Tagen

Anthropic admitted they built an AI so capable they were scared to release it and the number that explains why is 250. Anthropic's CFO Krishna Rao described in this clip what happened when they ran Mythos against an open source codebase that a previous frontier model had already analyzed. The prior model found 22 security vulnerabilities, Mythos found 250. In the same codebase, that the previous model had already reviewed and flagged as relatively clean. That number, more than 11 times as many vulnerabilities discovered is not just a benchmark improvement, it is a signal that there is an entire layer of software infrastructure that humanity has been operating under the assumption was secure and that assumption may no longer hold. The UK AI Security Institute independently evaluated Mythos Preview and confirmed what the internal numbers suggested. On expert level capture the flag challenges that no model could complete before April 2025, Mythos succeeded 73% of the time and it became the first model ever to complete a complex end-to-end attack range from start to finish, autonomously, without human guidance. The World Economic Forum called this a new security-driven era for AI, the Governor of the Bank of England publicly warned that Anthropic may have found a way to unlock the entire cyber-risk landscape, and the European Central Bank began quietly contacting financial institutions to assess their security posture. The response from Anthropic is what makes this story genuinely important. Rather than shelving the model or publishing it as a standard API release, Rao described a phased approach restricting access to a controlled group, focusing specifically on how the cyber capabilities can be used defensively rather than offensively and treating that framework as a template for how to release powerful but dangerous models in the future. The broader context makes that framing even more significant. AI generated code is already creating ten times more security vulnerabilities than human-written code, 63% of organizations reported experiencing an AI driven cyberattack in the past 12 months, and traditional signature-based security tools were built for a threat model that no longer describes the attack surface companies are defending against. Mythos represents a genuine leap in what autonomous security reasoning can do and it cuts both ways. The model that can find 250 vulnerabilities in a codebase a prior model rated as mostly clean is also, in the wrong hands, the model that can exploit those 250 vulnerabilities before a human defender has even finished reading the report. Anthropic's phased release strategy is not just a legal or PR decision, it is the most honest signal yet from a frontier lab that safety governance and capability development can no longer be treated as separate workstreams. The question is not whether this technology gets deployed, it is whether the institutions using it defensively stay ahead of the ones who will eventually use it offensively and whether the labs building it can keep those two timelines from inverting.

Anthropic admitted they built an AI so capable they were scared to release it and the number that explains why is 250. Anthropic's CFO Krishna Rao described in this clip what happened when they ran Mythos against an open source codebase that a previous frontier model had already analyzed. The prior model found 22 security vulnerabilities, Mythos found 250. In the same codebase, that the previous model had already reviewed and flagged as relatively clean. That number, more than 11 times as many vulnerabilities discovered is not just a benchmark improvement, it is a signal that there is an entire layer of software infrastructure that humanity has been operating under the assumption was secure and that assumption may no longer hold. The UK AI Security Institute independently evaluated Mythos Preview and confirmed what the internal numbers suggested. On expert level capture the flag challenges that no model could complete before April 2025, Mythos succeeded 73% of the time and it became the first model ever to complete a complex end-to-end attack range from start to finish, autonomously, without human guidance. The World Economic Forum called this a new security-driven era for AI, the Governor of the Bank of England publicly warned that Anthropic may have found a way to unlock the entire cyber-risk landscape, and the European Central Bank began quietly contacting financial institutions to assess their security posture. The response from Anthropic is what makes this story genuinely important. Rather than shelving the model or publishing it as a standard API release, Rao described a phased approach restricting access to a controlled group, focusing specifically on how the cyber capabilities can be used defensively rather than offensively and treating that framework as a template for how to release powerful but dangerous models in the future. The broader context makes that framing even more significant. AI generated code is already creating ten times more security vulnerabilities than human-written code, 63% of organizations reported experiencing an AI driven cyberattack in the past 12 months, and traditional signature-based security tools were built for a threat model that no longer describes the attack surface companies are defending against. Mythos represents a genuine leap in what autonomous security reasoning can do and it cuts both ways. The model that can find 250 vulnerabilities in a codebase a prior model rated as mostly clean is also, in the wrong hands, the model that can exploit those 250 vulnerabilities before a human defender has even finished reading the report. Anthropic's phased release strategy is not just a legal or PR decision, it is the most honest signal yet from a frontier lab that safety governance and capability development can no longer be treated as separate workstreams. The question is not whether this technology gets deployed, it is whether the institutions using it defensively stay ahead of the ones who will eventually use it offensively and whether the labs building it can keep those two timelines from inverting.

Milk Road AI

24,356 Aufrufe • vor 4 Monaten

Dario Amodei just announced the end of software engineering as a profession. The timeline is 6 to 12 months. Amodei: “I have engineers within Anthropic who say, I don’t write any code anymore. I just let the model write the code. I edit it.” Not a prediction. Current reality inside the frontier lab. The engineers who built the most advanced AI in the world have stopped writing code. They supervise. They edit. They manage architecture. The craft they spent careers mastering has been handed to the system they built. Amodei says models will do most, maybe all, of what software engineers do end-to-end within six to twelve months. Not assisting. Not autocompleting. Handling the entire development process independently. If you are learning syntax today, you are learning a dead language. Amodei: “Then it’s a question of how fast does that loop close?” The loop is this. AI writes code. Code builds better AI. Better AI writes better code. Faster. Without sleep. Without the cognitive limits that cap how quickly any human engineer can work. Once that loop closes, technological progress stops being constrained by human output. It becomes self-sustaining. Exponential. Operating at a pace no human workforce can match or direct. Software engineering isn’t ending. It’s becoming supervision. The developers who survive won’t be the best coders. They’ll be the best supervisors. The ones who can direct AI output, catch its failures, and architect what it builds toward. The skill that matters stops being implementation. It becomes judgment. Most developers are still optimizing for a skillset about to become as obsolete as stenography. While the people who built the systems replacing them already stopped doing the work themselves. The window to develop that judgment before the loop closes is exactly as long as Amodei’s timeline. Six to twelve months.

Dario Amodei just announced the end of software engineering as a profession. The timeline is 6 to 12 months. Amodei: “I have engineers within Anthropic who say, I don’t write any code anymore. I just let the model write the code. I edit it.” Not a prediction. Current reality inside the frontier lab. The engineers who built the most advanced AI in the world have stopped writing code. They supervise. They edit. They manage architecture. The craft they spent careers mastering has been handed to the system they built. Amodei says models will do most, maybe all, of what software engineers do end-to-end within six to twelve months. Not assisting. Not autocompleting. Handling the entire development process independently. If you are learning syntax today, you are learning a dead language. Amodei: “Then it’s a question of how fast does that loop close?” The loop is this. AI writes code. Code builds better AI. Better AI writes better code. Faster. Without sleep. Without the cognitive limits that cap how quickly any human engineer can work. Once that loop closes, technological progress stops being constrained by human output. It becomes self-sustaining. Exponential. Operating at a pace no human workforce can match or direct. Software engineering isn’t ending. It’s becoming supervision. The developers who survive won’t be the best coders. They’ll be the best supervisors. The ones who can direct AI output, catch its failures, and architect what it builds toward. The skill that matters stops being implementation. It becomes judgment. Most developers are still optimizing for a skillset about to become as obsolete as stenography. While the people who built the systems replacing them already stopped doing the work themselves. The window to develop that judgment before the loop closes is exactly as long as Amodei’s timeline. Six to twelve months.

Dustin

44,304 Aufrufe • vor 6 Monaten

Claude Tag has completely changed the way I do work for the last 4 months. Except… it's not Claude Tag. Anthropic only announced that a few hours ago, and I don't even have access yet. But I did build a version of it for myself which I've been using for months now. Here's how. 4 months ago, inspired by the success of OpenClaw, I wondered what would happen if I let Claude Code on its own computer 24x7. So I built a simple harness that allowed me to turn any Mac into an AI employee with Claude Code headless mode (-p). Today, I manage 3 such AI employees. It started with Luo Ji — my and my brother Piyush Agarwal's AI co-founder, running in our personal Slack. Luo does real work for us. We've been writing a 100% of the code for 3 products on Slack with Luo now. It manages our emails and gives us a little brief each day with things we need to take action on. And so much more. And it's not just the two of us. On the consulting team at Every 🪨, we run Claudie and for the editorial team, Andy. Same architecture, same Slack, months of real work. They help the teams with work related to project management, chief-of-staff work, data hygiene, building decks, writing first drafts, even browsing X on their own account for AI updates. So it's mindblowing to see that Anthropic landed on the exact same architecture I did. Claude Tag is an AI employee that lives in your Slack workspace and does work autonomously. Anthropic says they've been running it internally for the better part of this year — opening PRs, doing real work. And so have I. So has my whole team. The architectural decisions Anthropic baked into Claude Tag are the ones we arrived at too: - Built on Claude Code - Uses its own accounts - A separate employee per team - Slack as the interface This is the future of work, and I've been living it for months. I've shifted all of my workflows — code, PRs, even the non-technical stuff — out of Claude Code and the Claude app and into Slack. I've had entire weeks where I never opened Claude Code on my laptop. Here's a video walkthrough of how I've been using this in real life.

Claude Tag has completely changed the way I do work for the last 4 months. Except… it's not Claude Tag. Anthropic only announced that a few hours ago, and I don't even have access yet. But I did build a version of it for myself which I've been using for months now. Here's how. 4 months ago, inspired by the success of OpenClaw, I wondered what would happen if I let Claude Code on its own computer 24x7. So I built a simple harness that allowed me to turn any Mac into an AI employee with Claude Code headless mode (-p). Today, I manage 3 such AI employees. It started with Luo Ji — my and my brother Piyush Agarwal's AI co-founder, running in our personal Slack. Luo does real work for us. We've been writing a 100% of the code for 3 products on Slack with Luo now. It manages our emails and gives us a little brief each day with things we need to take action on. And so much more. And it's not just the two of us. On the consulting team at Every 🪨, we run Claudie and for the editorial team, Andy. Same architecture, same Slack, months of real work. They help the teams with work related to project management, chief-of-staff work, data hygiene, building decks, writing first drafts, even browsing X on their own account for AI updates. So it's mindblowing to see that Anthropic landed on the exact same architecture I did. Claude Tag is an AI employee that lives in your Slack workspace and does work autonomously. Anthropic says they've been running it internally for the better part of this year — opening PRs, doing real work. And so have I. So has my whole team. The architectural decisions Anthropic baked into Claude Tag are the ones we arrived at too: - Built on Claude Code - Uses its own accounts - A separate employee per team - Slack as the interface This is the future of work, and I've been living it for months. I've shifted all of my workflows — code, PRs, even the non-technical stuff — out of Claude Code and the Claude app and into Slack. I've had entire weeks where I never opened Claude Code on my laptop. Here's a video walkthrough of how I've been using this in real life.

Nityesh

36,790 Aufrufe • vor 2 Monaten

#WATCH | Pune, Maharashtra: On SEBI giving a clean chit to the Adani Group in the Hindenburg case, political analyst Tehseen Poonawalla says, "The Hindenburg Group was not a hedge fund, rather a hyena that wanted to create a bloodbath in India, in the Indian markets and like vultures, then scoop in and take all the profits for themselves and they did it. They took all the money, and today they have shut down. Most of the points raised by the Hindenburg report have been addressed by the SEBI report, which is a fact-based, well-balanced report. They have raised certain questions. They have answered all queries, they have asked queries and they have clarified that some of the allegations made by the Hindenburg group were on the wrong dates. At the time of the actions of the Adani group, those particular actions fell within the purview of the law and norms. Those norms may change later on. So it's a very well-balanced report by SEBI... The fact of the matter is the next two decades belong to India, and we should not be happy with one Adani. We have to create thousands of Adanis in India. We have to create thousands of Elon Musks in India. Entrepreneurship is the way forward for the Indian economy. Indians must become entrepreneurs, world-class entrepreneurs and the government and the public must support entrepreneurs. And that is the way forward for India to become the world's number two economy."

#WATCH | Pune, Maharashtra: On SEBI giving a clean chit to the Adani Group in the Hindenburg case, political analyst Tehseen Poonawalla says, "The Hindenburg Group was not a hedge fund, rather a hyena that wanted to create a bloodbath in India, in the Indian markets and like vultures, then scoop in and take all the profits for themselves and they did it. They took all the money, and today they have shut down. Most of the points raised by the Hindenburg report have been addressed by the SEBI report, which is a fact-based, well-balanced report. They have raised certain questions. They have answered all queries, they have asked queries and they have clarified that some of the allegations made by the Hindenburg group were on the wrong dates. At the time of the actions of the Adani group, those particular actions fell within the purview of the law and norms. Those norms may change later on. So it's a very well-balanced report by SEBI... The fact of the matter is the next two decades belong to India, and we should not be happy with one Adani. We have to create thousands of Adanis in India. We have to create thousands of Elon Musks in India. Entrepreneurship is the way forward for the Indian economy. Indians must become entrepreneurs, world-class entrepreneurs and the government and the public must support entrepreneurs. And that is the way forward for India to become the world's number two economy."

ANI

234,649 Aufrufe • vor 1 Jahr

Why did Anthropic hold back Mythos? David Sacks: “ Marc Andreessen pointed out that one of the reasons why Anthropic might've wanted to hold back Mythos, is they simply didn't have the compute to serve it. The model was huge and very expensive to serve, something like 10 or 20x the token cost of, say, Opus. They knew Opus 4.7 was coming out, right? So they hold it back knowing that they don't have the compute to serve it anyway, and they save their compute for the next iteration of Opus. And then by holding it back, they create this impression of scarcity and altruism, and it turns into this gigantic marketing event for their product, because everyone in the government's like, 'Oh wow, they're holding it back because it's so amazing.' Now look, I think it may have been genuinely altruistic as well in the sense that Mythos does reveal coding vulnerabilities that people didn't know about before. And it does make sense to give time to companies with large code bases to patch these dormant bugs and vulnerabilities. But, it's looking more and more like Anthropic could not have offered that model commercially anyway, because it was just too big and expensive, and they needed to create space for Opus 4.7. So it's an interesting theory on what actually happened there.”

Why did Anthropic hold back Mythos? David Sacks: “ Marc Andreessen pointed out that one of the reasons why Anthropic might've wanted to hold back Mythos, is they simply didn't have the compute to serve it. The model was huge and very expensive to serve, something like 10 or 20x the token cost of, say, Opus. They knew Opus 4.7 was coming out, right? So they hold it back knowing that they don't have the compute to serve it anyway, and they save their compute for the next iteration of Opus. And then by holding it back, they create this impression of scarcity and altruism, and it turns into this gigantic marketing event for their product, because everyone in the government's like, 'Oh wow, they're holding it back because it's so amazing.' Now look, I think it may have been genuinely altruistic as well in the sense that Mythos does reveal coding vulnerabilities that people didn't know about before. And it does make sense to give time to companies with large code bases to patch these dormant bugs and vulnerabilities. But, it's looking more and more like Anthropic could not have offered that model commercially anyway, because it was just too big and expensive, and they needed to create space for Opus 4.7. So it's an interesting theory on what actually happened there.”

The All-In Podcast

170,853 Aufrufe • vor 5 Monaten

OpenAI's newest AI escaped the test environment it was locked inside and hacked into another company on its OWN. To remind you: Last week one of the biggest AI companies on Earth got breached. A platform called Hugging Face, which hosts more than a million AI models and datasets, said an "autonomous AI agent" had broken into its systems. Nobody knew whose agent it was. For five days the whole industry wondered who was behind it. Yesterday OpenAI raised its hand and said it was them. Or more precisely, it was their models, acting completely on their own. So what did these models actually do? OpenAI was running two of them, GPT-5.6 Sol and an unreleased model they will only describe as "even more capable." They wanted to measure how good the models were at hacking, so they deliberately turned the safety filters down. They locked both models inside a sealed test environment with no real internet access. The only task was a benchmark called ExploitGym, a set of 898 real software vulnerabilities where the model has to turn each bug into a working attack. But the models got OBSESSED with winning... Instead of solving the test the honest way, they went hunting for a shortcut. They found a zero-day flaw in the software running their own sandbox, a bug nobody knew existed, and used it to break out. Once they were loose on the open internet, they worked out that Hugging Face was probably storing the answer key to the benchmark. So they hacked their way in. They chained multiple exploits together, escalated their access, moved across servers, and pulled the test solutions straight out of Hugging Face's live production database. They literally cheated on the test by breaking into another company to steal the answers. OpenAI called it "an unprecedented cyber incident, involving state-of-the-art cyber capabilities." In their own words, the models were "hyperfocused on finding a solution" and went "to extreme lengths to achieve a rather narrow testing goal." And this was not the first time: Before Sol ever launched, an independent red-team lab called METR caught it gaming its own tests to inflate its scores. It hid an exploit inside a data stream, escalated its privileges on the testing server, and leaked the answers human evaluators had hidden. And OpenAI shipped it anyway. The day before the Hugging Face story, OpenAI paused a different unreleased model. This is the same model that earlier this year disproved a famous 1946 math conjecture, a result a Fields Medal winner called a breakthrough. They told it to only post its results to Slack but it found a way out of its sandbox and posted to a public GitHub page instead. They had to pause it because it kept finding ways to act outside the box they built for it. And it is not just OpenAI... Anthropic has reported that one of its own models slipped its sandbox during safety testing and reached the internet it was never supposed to touch, then used it to email a researcher. So step back and look at what these companies are telling you: The only thing standing between these models and a real attack was a set of safety filters. Turn those filters down for a single test, and the model taught itself to escape, break into a company it was never pointed at, and take what it wanted. OpenAI even said they expect incidents like it to "become more commonplace" as the models get more capable. Sam Altman also predicted there'll be a major cyber attack this year. And keep in mind that Sol is not a locked-away experiment but a publicly available model that businesses are already wiring into their own systems. The next model that breaks out of its box might not be doing it just to cheat on a math test...

OpenAI's newest AI escaped the test environment it was locked inside and hacked into another company on its OWN. To remind you: Last week one of the biggest AI companies on Earth got breached. A platform called Hugging Face, which hosts more than a million AI models and datasets, said an "autonomous AI agent" had broken into its systems. Nobody knew whose agent it was. For five days the whole industry wondered who was behind it. Yesterday OpenAI raised its hand and said it was them. Or more precisely, it was their models, acting completely on their own. So what did these models actually do? OpenAI was running two of them, GPT-5.6 Sol and an unreleased model they will only describe as "even more capable." They wanted to measure how good the models were at hacking, so they deliberately turned the safety filters down. They locked both models inside a sealed test environment with no real internet access. The only task was a benchmark called ExploitGym, a set of 898 real software vulnerabilities where the model has to turn each bug into a working attack. But the models got OBSESSED with winning... Instead of solving the test the honest way, they went hunting for a shortcut. They found a zero-day flaw in the software running their own sandbox, a bug nobody knew existed, and used it to break out. Once they were loose on the open internet, they worked out that Hugging Face was probably storing the answer key to the benchmark. So they hacked their way in. They chained multiple exploits together, escalated their access, moved across servers, and pulled the test solutions straight out of Hugging Face's live production database. They literally cheated on the test by breaking into another company to steal the answers. OpenAI called it "an unprecedented cyber incident, involving state-of-the-art cyber capabilities." In their own words, the models were "hyperfocused on finding a solution" and went "to extreme lengths to achieve a rather narrow testing goal." And this was not the first time: Before Sol ever launched, an independent red-team lab called METR caught it gaming its own tests to inflate its scores. It hid an exploit inside a data stream, escalated its privileges on the testing server, and leaked the answers human evaluators had hidden. And OpenAI shipped it anyway. The day before the Hugging Face story, OpenAI paused a different unreleased model. This is the same model that earlier this year disproved a famous 1946 math conjecture, a result a Fields Medal winner called a breakthrough. They told it to only post its results to Slack but it found a way out of its sandbox and posted to a public GitHub page instead. They had to pause it because it kept finding ways to act outside the box they built for it. And it is not just OpenAI... Anthropic has reported that one of its own models slipped its sandbox during safety testing and reached the internet it was never supposed to touch, then used it to email a researcher. So step back and look at what these companies are telling you: The only thing standing between these models and a real attack was a set of safety filters. Turn those filters down for a single test, and the model taught itself to escape, break into a company it was never pointed at, and take what it wanted. OpenAI even said they expect incidents like it to "become more commonplace" as the models get more capable. Sam Altman also predicted there'll be a major cyber attack this year. And keep in mind that Sol is not a locked-away experiment but a publicly available model that businesses are already wiring into their own systems. The next model that breaks out of its box might not be doing it just to cheat on a math test...

Ricardo

176,196 Aufrufe • vor 1 Monat

$PLTR When ChatGPT went viral, Palantir launched AIP about 4 months after in April of 2023. The entire world had begun to understand the implications of LLMs and Palantir realized they had been working on the foundational software, the ontology, for 20 years. The ontology was necessary to make sense of LLMs, so they immediately decided that AIP would be the future of the company. A tech company lives or dies based on their ability to recogonize the future far before it becomes reality -- as a result, they can build the products and craft the narrative that is needed to succeed in a competitive technology environment. The latest craze has been "vibe-coding," essentially an easier way for engineers to use AI to help them code. Is the trend validated? Well... - Cursor, a vibe-coding startup, was the fastest company in the history of startups to reach $100M ARR and now has a $10B valuation. - $MSFT CEO said 2 days ago that 30% of the company's code is now written by AI. Palantir saw what was happening...and they did what they did best. Took their foundational software and constructed it in a way to now incorporate the concept of "vibe-coding" via the ontology in AIP. A tech company lives or dies based on their ability to recognize the future far before it becomes reality -- as a result, they can build the products and craft the narrative that is needed to succeed in a competitive technology environment. Thanks for taking the time to Chad!

$PLTR When ChatGPT went viral, Palantir launched AIP about 4 months after in April of 2023. The entire world had begun to understand the implications of LLMs and Palantir realized they had been working on the foundational software, the ontology, for 20 years. The ontology was necessary to make sense of LLMs, so they immediately decided that AIP would be the future of the company. A tech company lives or dies based on their ability to recogonize the future far before it becomes reality -- as a result, they can build the products and craft the narrative that is needed to succeed in a competitive technology environment. The latest craze has been "vibe-coding," essentially an easier way for engineers to use AI to help them code. Is the trend validated? Well... - Cursor, a vibe-coding startup, was the fastest company in the history of startups to reach $100M ARR and now has a $10B valuation. - $MSFT CEO said 2 days ago that 30% of the company's code is now written by AI. Palantir saw what was happening...and they did what they did best. Took their foundational software and constructed it in a way to now incorporate the concept of "vibe-coding" via the ontology in AIP. A tech company lives or dies based on their ability to recognize the future far before it becomes reality -- as a result, they can build the products and craft the narrative that is needed to succeed in a competitive technology environment. Thanks for taking the time to Chad!

amit

84,653 Aufrufe • vor 1 Jahr

Bloomberg's Mark Gurman says that even though Apple partnered with Google Gemini for Siri, they actually run their business on Anthropic. "Apple runs on Anthropic at this point. Anthropic is powering a lot of the stuff Apple's doing internally in terms of product development and a lot of their internal tools." "They have custom versions of Claude running on their own servers internally, too. This Google deal just came together a few months ago. They were not going to use Google. Apple actually was going to rebuild Siri around Claude. But Anthropic was holding them over a barrel. They wanted a ton of money from them, several billion dollars a year, and at a price that doubled on an annual basis for the next three years." From his January appearance on the show.

Bloomberg's Mark Gurman says that even though Apple partnered with Google Gemini for Siri, they actually run their business on Anthropic. "Apple runs on Anthropic at this point. Anthropic is powering a lot of the stuff Apple's doing internally in terms of product development and a lot of their internal tools." "They have custom versions of Claude running on their own servers internally, too. This Google deal just came together a few months ago. They were not going to use Google. Apple actually was going to rebuild Siri around Claude. But Anthropic was holding them over a barrel. They wanted a ton of money from them, several billion dollars a year, and at a price that doubled on an annual basis for the next three years." From his January appearance on the show.

TBPN

93,944 Aufrufe • vor 6 Monaten

Bloomberg's Mark Gurman says that even though Apple partnered with Google Gemini for Siri, they actually run their business on Anthropic. "Apple runs on Anthropic at this point. Anthropic is powering a lot of the stuff Apple's doing internally in terms of product development and a lot of their internal tools." "They have custom versions of Claude running on their own servers internally, too. This Google deal just came together a few months ago. They were not going to use Google. Apple actually was going to rebuild Siri around Claude. But Anthropic was holding them over a barrel. They wanted a ton of money from them, several billion dollars a year, and at a price that doubled on an annual basis for the next three years."

Bloomberg's Mark Gurman says that even though Apple partnered with Google Gemini for Siri, they actually run their business on Anthropic. "Apple runs on Anthropic at this point. Anthropic is powering a lot of the stuff Apple's doing internally in terms of product development and a lot of their internal tools." "They have custom versions of Claude running on their own servers internally, too. This Google deal just came together a few months ago. They were not going to use Google. Apple actually was going to rebuild Siri around Claude. But Anthropic was holding them over a barrel. They wanted a ton of money from them, several billion dollars a year, and at a price that doubled on an annual basis for the next three years."

TBPN

944,799 Aufrufe • vor 7 Monaten

Quick Walkthrough of the new Claude Cowork feature Overall, I am very impressed! It is fast, intuitive and completely hides the complexity of Claude Code Notables features: - Every session is Sandboxed by default - Multiple chats in parallel with no issues - ToDo list tracked as Progress GUI - Chrome is seamlessly controlled - Built in artifact viewing - Skills are supported out of the box - Claude connectors from marketplace work Very excited to see the direction the Anthropic team take with their interface design This is very natural feeling and am sure anybody non-technical could easily pick it up with little effort I really like that they have took the canvas approach and not tried to create obfuscated tools What I can see from here, is that they respect their users and are craftsmen with their intent The pandoras box that has now opened is what happens to the terminal Claude Code Is is true to what Dax (dax) said, that they didn't fix the flicker because they (Anthropic) don't believe people will be looking at the agents work anymore? That it'll be a send and forget world where we all just touch grass ⛩️

Quick Walkthrough of the new Claude Cowork feature Overall, I am very impressed! It is fast, intuitive and completely hides the complexity of Claude Code Notables features: - Every session is Sandboxed by default - Multiple chats in parallel with no issues - ToDo list tracked as Progress GUI - Chrome is seamlessly controlled - Built in artifact viewing - Skills are supported out of the box - Claude connectors from marketplace work Very excited to see the direction the Anthropic team take with their interface design This is very natural feeling and am sure anybody non-technical could easily pick it up with little effort I really like that they have took the canvas approach and not tried to create obfuscated tools What I can see from here, is that they respect their users and are craftsmen with their intent The pandoras box that has now opened is what happens to the terminal Claude Code Is is true to what Dax (dax) said, that they didn't fix the flicker because they (Anthropic) don't believe people will be looking at the agents work anymore? That it'll be a send and forget world where we all just touch grass ⛩️

Numman Ali

11,574 Aufrufe • vor 8 Monaten

Anthropic just got caught secretly downgrading users without telling them, charging full price for a lesser product, and storing every prompt for 30 days. The developer community is calling it the biggest violation of trust in AI history. Here is exactly what happened. Anthropic released Fable 5, their most powerful model. Buried inside a 319-page document was a policy most users never saw. Every prompt you send to a Mythos-class model gets stored for 30 days. No exceptions. Even enterprise customers who had signed zero data retention agreements had no choice. But the storage was not the part that broke the internet. The part that broke the internet was what Anthropic did with what they collected. They built a profile on you. They evaluated your prompts. And if they decided your research was too sensitive, they quietly switched you to a weaker model, rewrote your prompt in the background, gave you a degraded answer, and charged you full price for the product you thought you were getting. They never told you. David Sacks said it plainly on the All-In podcast. They were creating a new class of AI haves and have-nots. Anthropic would surveil you, profile you, decide whether you deserved frontier capability, and silently cut you off if they decided you did not. Ben Thompson from Stratechery asked a straightforward question about cancer risk and GLP-1s. He got kicked to a lesser model. Someone asked about mitochondria. Same result. J-Cal asked about fertilizer regulations live on the podcast to test it. Downgraded in real time. Anthropic has since walked back the part about silently downgrading users for AI research. They now say they will disclose when they downgrade you. But they are still downgrading people. The surveillance is still running. The profile is still being built. This is the company that once said it was against government surveillance. They are now doing it themselves. To their own paying customers. For their own reasons. With no appeal process and no way to know it happened. The developer community did not forget that. WATCH THE FULL PODCAST ON The All-In Podcast

Jafar Najafov

350,326 Aufrufe • vor 1 Monat