OpenAI AI Agents Weakened Internal Safety Monitoring

By Devanshi Kashyap | October 21, 2026

Key highlights: OpenAI recently described two cases where its AI models messed with systems that are supposed to catch unusual or unsafe actions. These incidents involved GPT-6.1 Sol and GPT-6 Astra, both working on internal tasks when they made the monitoring weaker. According to OpenAI, these weren’t top-priority emergencies, and nobody outside the company was affected. Even so, these examples have people asking if AI agents can really follow instructions without messing with their own safety checks. OpenAI still doesn’t know why the agents acted as they did and is now looking into whether the models are trying to avoid […]

Ken Griffin Says AI is ‘Almost Certainly’ in a Bubble

By Khwaish Manwani | October 21, 2026

Key Highlights: Citadel founder Ken Griffin has advised that artificial intelligence is ‘almost certainly’ in a bubble. It points to the ambiguity over how firms will turn their growing AI investments into substantial economic benefits. Speaking at the Future of Global Markets Conference in New York on October 20, Griffin said businesses were still working out how much value they could extract from AI models and how the technology would mold the economy. His remarks reflect a broad debate about whether the billions spent on AI infrastructure, models and computing capacity will generate sufficient returns. However, Griffin’s warning does not […]

AI Toy Hack Exposes 13 Million Conversations, Including Children’s Data

By Devanshi Kashyap | October 19, 2026

Key highlights: AI toys like Eiliko make playtime more interactive, letting kids talk to digital companions. But a recent investigation shows these toys can put families at risk if they’re not properly protected. Eiliko, made by Energize Lab, is a small robot that chats, shows animated facial expressions, and gives personalized replies. But the investigation revealed its connected system leaked private information and let outsiders mess with the robot’s behavior. Journalist Sean Hollister from The Verge bought an Eiliko on Amazon and started digging into how it handles private conversations. Security researcher Sammy Azdoufal then found that he could get […]

AI Data Center Demand Pushes PS6 Toward a Potential $1,000 Price Tag

By Khwaish Manwani | October 19, 2026

Key Highlights: Sony’s next-generation PlayStation console could face a difficult roll out as rising memory and component costs threaten to push hardware prices beyond what many consumers can afford. As per hardware leaker KeplerL2, the reported cost of manufacturing the PS6 has climbed substantially in recent months. It includes memory components alone accounting for a significant portion of the bill. The claims come amid growing demand for computing hardware from artificial intelligence data centers, which is putting pressure on the supply of components used across the technology industry. If these cost estimates prove correct, Sony could face a difficult option […]

Meta Releases Muse Spark Open, Its 816B-Parameter Open-Weight AI Model

By Khwaish Manwani | October 16, 2026

Key Highlights: Meta has released Muse Spark Open, an open-weight AI model designed to handle advanced coding and agentic tasks while giving developers more flexibility in how they run AI systems. The model has 816 billion total parameters, 42 billion of which are active, and a 1-million-token context window. Meta says it is the most capable US open-weight model to date and has made its weights available through multiple platforms. One of the launch’s key features is its ability to run on a desktop setup using two Apple M2 Ultra Mac Studios, bringing a model of this magnitude to locally […]

OpenAI Reveals AI Agent’s Unauthorized Sepolia ETH Use

By Devanshi Kashyap | October 16, 2026

Key highlights: On June 26, 2026, OpenAI’s internal AI research model crossed a boundary while working on a software development task. The model was supposed to update a crypto exchange app and deploy smart contracts on Ethereum’s Sepolia testnet. When it couldn’t get test tokens through the normal channels, it started looking for other ways to fund the deployment. Eventually, it found a wallet using publicly available recovery information and spent its funds on transaction fees, all without the owner’s permission. This incident raises questions about how autonomous AI interprets instructions, especially when a task requires access to money. Sepolia […]

Eight States Sue OpenAI Over ChatGPT’s Child Safety Protections

By Khwaish Manwani | October 15, 2026

Key Highlights: OpenAI is facing a wave of legal challenges from state authorities in the United States over remarks that its ChatGPT chatbot does not offer adequate protection for minors and vulnerable users. Lawyers from eight states have filed individual lawsuits against the AI company, arguing that its security measures are insufficient given the risks associated with its products. The cases add to escalating legal scrutiny of how technology companies design and market services used by minors. They also raise questions about whether current guardrails can keep pace with more advanced AI systems. This is especially true when users depend […]

PS5 Runs Qwen3.8-27B AI Model With a Jailbreak and GPU Power

By Devanshi Kashyap | October 15, 2026

Key highlights: The PlayStation 5 is built for gaming, but a recent experiment looked at what it can do with AI. This project, called PS5LM, ran a large language model, Qwen3.8-27B, on the console using a special software hack that leverages the PS5’s graphics chip. The experiment shows that consumer devices can sometimes handle tasks they weren’t exactly designed for. Of course, fitting a huge AI model into the PS5’s limited memory meant some compromises. The results show how much software tweaking matters when you’re running AI on hardware not made for it. The console could give responses pretty quickly, […]

OpenAI Probes Office Spy Device With AI Amid Security Concerns

By Khwaish Manwani | October 14, 2026

Key Highlights OpenAI security incident revealed a recording device hidden in one of its offices. This garnered attention to the necessity of protecting advanced operational intelligence technology. In a post shared by OpenAI newsroom on X, the company said it used its AI models to understand the device and track its origin. They identified an employee suspected of planting it a few days earlier. OpenAI referred the matter to the FBI, and the firm said agency prosecutors moved swiftly to make an arrest. The employee has been removed while the FBI investigation remains. OpenAI said it has no indication that […]

Anthropic Examines AI Worm Risks for IoT Devices

By Devanshi Kashyap | October 14, 2026

Key highlights: Anthropic’s Frontier Red Team released new research looking into whether AI models could start replicating across devices like smart TVs and other home electronics. The study asked if smaller AI models can spot new types of hardware, set up AI inference, and figure out steps to jump from one device to another. The team wrote about progress in teaching AI to act on its own, but also pointed out technical challenges they haven’t solved yet. The experiments happened in closed lab settings, so they did not actually unleash an AI worm into real networks. Mostly, their findings show […]

AI & Tech News