Ad
Skip to content

Maximilian Schreiner

Maximilian Schreiner is Managing Editor at THE DECODER and has covered artificial intelligence for close to a decade, from new models and tools to their impact on businesses, media, and society. He brings that expertise to heise KI PRO, where he makes AI accessible for professionals in the DACH region - clear, critical, and practical. With a background in philosophy, one question keeps driving him: whether machines can truly think, or are just getting better at faking it.

Former OpenAI researcher says he found Nvidia Blackwell fingerprints hidden inside six Chinese AI models

Former OpenAI researcher Will DePue says he has produced the first technical evidence that DeepSeek, Moonshot AI, MiniMax, and Ant Group are using Nvidia’s latest Blackwell GPUs despite US export controls. His forensic method reads characteristic chip signatures directly from published model weights. He estimates the hardware behind the five training runs cost more than half a billion dollars.

Read full article about: Anthropic's IPO is six times oversubscribed and on track to become the largest public offering in history

Anthropic is on track for the largest IPO ever, with orders exceeding available shares by a factor of six. That's according to Bloomberg. The Claude developer and its co-founders aim to raise up to $64.4 billion through the public offering and another $26 billion via a private placement, which would surpass even SpaceX's IPO. The target valuation sits at roughly $1.4 trillion, with existing investors getting priority in share allocation.

The company's financials likely fuel much of the demand. Anthropic's annualized revenue climbed to $81 billion by late October, up from $71 billion in September. Nvidia, whose investors have grown more cautious since CEO Jensen Huang was ousted over an alleged espionage scandal, is itself buying $10 billion in Anthropic stock through the private placement. Goldman Sachs and Morgan Stanley are managing the Nasdaq listing under the ticker ANT, and CEO Dario Amodei will pitch investors alongside JPMorgan CEO Jamie Dimon in a Friday livestream.

Read full article about: Meta's Muse now calls businesses for you, but many companies might just hang up on the AI

Meta's AI agent Muse can now place calls to US businesses on behalf of its users. The feature, tested in September, is available to all paying Muse subscribers in the US, turning what Google Duplex demoed to widespread amazement in 2018 into a standard capability of always-on AI agents.

Meta AI chief Alexandr Wang said the agent, with more than 10 million users, handles tedious calls to cable companies, the DMV, and restaurants, and will soon get a free email address. Google's Gemini, startup Instinct, and xAI's Grok Bot can already place calls as well, while OpenAI's ChatGPT Dots and Anthropic's Claude Duet still cannot.

Many businesses, however, hang up the moment they detect an AI on the line. That problem forced Meta to temporarily test human operators for Muse in September, echoing Google's experience with Duplex, where on launch one in four calls was handled by a human. Synthetic voices keep getting more realistic, making that detection increasingly difficult. Users are drawn to the feature largely because the AI will sit on hold indefinitely and work through complex phone menus without complaint.

Read full article about: OpenAI enters the cybersecurity market with two Daybreak products

OpenAI has launched two security products under its Daybreak brand. The first, Daybreak Red Team, is available in Early Access and uses AI agents to automatically probe web applications for vulnerabilities by analyzing source code and app context to map out attack paths. The tool also verifies reported security flaws from bug bounty programs or code scanners and tests patches. Nancy Wang, head of Cyber-Applied AI, said the tool is designed to deliver penetration testing at the speed of automated continuous scanning while matching the depth of human testers.

The second product, Daybreak Defense Factory, automates the defensive side. AI agents scan a company's codebase for vulnerabilities and generate patches, delivering ready-to-deploy pull requests with fixes and accompanying tests for developer approval. Both products run on OpenAI's latest cyber model, GPT-6 Cyber.

OpenAI’s data suggests that frontier models are hiding their thoughts because developers optimize for capability and compute

OpenAI’s safety team systematically investigated why the chains of thought in frontier models like GPT-6 Astra are increasingly evading oversight. Four training steps drive the erosion, including architecture choices and reinforcement learning. The analysis shows that the loss of transparency is not a law of nature but also reflects tradeoffs around capability and cost.

Read full article about: Jensen Huang resigns as Nvidia CEO after espionage scandal wipes out $1.4 trillion in market value

Jensen Huang resigned as CEO, president, and board member of Nvidia effective immediately, according to a statement on November 12, 2026. The move came after the Wall Street Journal reported that the board was planning to fire Huang following revelations about two extramarital affairs, one of them with a suspected Chinese spy.

Since a New York Times report on October 29 about possible ties to Chinese intelligence, Nvidia has shed roughly $1.4 trillion in market value. CFO Colette Kress is taking over on an interim basis while the board searches for a permanent successor. Harvey C. Jones will serve as the first independent board chair and stressed that business operations continue without disruption.

Replacing Huang will be difficult because Nvidia was built entirely around him. He secured the company's market position through a complex web of alliances with financiers and roughly $300 billion in guarantees for smaller customers. His departure could unravel that structure, which he used to keep major tech rivals at bay and support large parts of the AI industry.

Major customers like Microsoft, Amazon, and Alphabet, along with AI labs like OpenAI and Anthropic, depend heavily on Nvidia's chips, but the company's 75 percent profit margin is pushing them to develop their own, cheaper alternatives. Huang also left no clear succession plan, with more than 60 employees reporting directly to him, leaving both internal and external candidates starting from a weakened position.

ElevenLabs Scribe v3 Live uses conversational context to stop voice agents from cutting off callers mid-sentence

ElevenLabs has released Scribe v3 and Scribe v3 Live, two new speech recognition models designed to transcribe accurately through background noise and crosstalk. Benchmarks from Artificial Analysis confirm that Scribe v3 narrowly beats competitors like OpenAI, Google, and Microsoft on accuracy. That precision comes at a cost, though. Rivals like Microsoft and xAI run many times faster and charge roughly half the price.

US data brokers are selling half a billion dollars in RL training environments to Chinese AI labs

US data brokers like Mercor, Surge AI, and AfterQuery are selling roughly half a billion dollars’ worth of AI training data to Chinese labs, according to a ChinaTalk investigation. That includes the same RL environments previously built for OpenAI or Anthropic. At least nine firms are involved, and export controls barely apply.