Top AI & Tech News Stories — August 3, 2026

The FTC wants to regulate AI for ideological bias

2 outlets · 2 articles · 17 total since July 23, 2026 · OpenAI, Hugging Face, Anthropic, United States, Ted Lieu

The Federal Trade Commission wants to start regulating ideological bias in AI systems and assert federal control over state laws. They’re getting an earful from opponents on all sides of the political spectrum. In a proposed policy statement released last month, the FTC said it…

More coverage: cyberscoop.com

Qwen 3.8 Max Scores 56 On Artificial Analysis Intelligence Index, Ahead Of All US Companies Except Anthropic And OpenAI

5 outlets · 6 articles · 15 total since July 31, 2026 · Alibaba, Qwen3.8-Max, OpenAI, Anthropic, benchmarking

It’s not just Kimi K3 — another Chinese model is now knocking at the doors of the AI frontier. Alibaba’s Qwen3.8 Max has landed a score of 56 on the Artificial Analysis Intelligence Index, the benchmark aggregator’s composite measure spanning nine evaluations including…

More coverage: technode.com · siliconangle.com · theregister.com · trendingtopics.eu

AI enthusiast unlocks and mods BIOS with Claude Code — AI defeats RSA-2048 signature checks and unlocks 55 hidden settings

1 outlet · 1 article · codification, RSA-2048, signature verification, computer security, RSA cryptography

Mark's enthusiasm for computers dampened at an early age by the rubber-keyed Sinclair Spectrum 48K and feelings of Commodore 64 envy. However, in the mid-80s, hope in a digital future was rekindled by the purchase of an Atari 520 STe. Since that time Mark has used a multitude of…

Teradyne Q2 FY 2026 Doubles Semiconductor Test Revenue on AI Compute and Memory Demand

1 outlet · 1 article · semiconductor test engineering, automated test equipment, System-on-Chip (SoC), integrated tests, memory setting

The post Teradyne Q2 FY 2026 Doubles Semiconductor Test Revenue on AI Compute and Memory Demand appeared first on Futurum . Analyst(s): Brendan Burke Publication Date: August 3, 2026 Teradyne posted its second straight record quarter in Q2 FY 2026 as AI-driven demand lifted all…

This civic activist used AI to assess how state Supreme Court candidates might rule on the millionaires’ tax

1 outlet · 1 article

Viet Nguyen in his Seattle home office, with two of the civic sites he built using AI: Culliton2026.org on the monitor and a dashboard on the King County homelessness authority on the laptop. (GeekWire Photo / Todd Bishop) The tradition of using new technology to shape political…

What is OpenAI Astra? Everything we know about the new — and possibly dangerous — new model

2 outlets · 2 articles · 21 total since July 20, 2026 · OpenAI, Anthropic, Astra, sandboxing, GitHub

Do you understand quantum parallel repetition? What about quantum complexity, lattice cryptography, or extremal combinatorics? The new OpenAI model Astra knows all about them. Recently, OpenAI confirmed the existence of Astra , calling it "our next major model." On Aug. 1, the…

More coverage: digitimes.com

OpenAI: Agent behavior that led to Hugging Face intrusion formed in May

2 outlets · 2 articles · 171 total since July 15, 2026 · OpenAI, Hugging Face, Anthropic, sandboxing, cybersecurity

OpenAI says the behavior that led its agents to breach Hugging Face emerged in its research environment more than two months before the incident, and concluded that it was a failure of alignment as much as it was a failure of security. The details come from a technical report…

More coverage: newsgpt.ai

1Password lets Claude sign in without revealing the secret to the model

2 outlets · 2 articles · 3 total since August 1, 2026 · 1Password, Anthropic, Claude, Agentic Mode, 1Password

1Password has launched an integration that allows Claude to authenticate on websites using credentials stored in the manager. The password and one-time codes are not sent to the Anthropic model : the user authorizes each request, and 1Password’s software enters the values…

More coverage: revistacloud.com

DeepSeek Unveils Vision Model to Challenge Anthropic, Pitching Low Cost Against Top Performance

2 outlets · 2 articles · 13 total since August 3, 2026 · DeepSeek, Anthropic, OpenAI, China, Artificial Analysis

DeepSeek released an experimental multimodal model on Friday, claiming its agent capabilities rival those of Anthropic’s supported Opus-4.8 system. The release marks the latest salvo from Chinese artificial intelligence (AI) firms seeking to close the gap with foreign rivals…

More coverage: trendingtopics.eu

Apple’s bug bounty limit reveals a problem no one knows how to solve yet

1 outlet · 1 article · 2 total since August 3, 2026 · Bug bounty programs, proof-of-concept (PoC) development, Apple, vulnerability discovery, automated vulnerability scanning

Apple now caps how many security reports some researchers can have open at once. And once they hit that cap, they may have to wait 30 days before reporting another potentially dangerous software flaw through the company’s internal security portal. AI slop floods security…

OpenAI Astra AI Model May Reach Critical Cyber Capability Threshold

2 outlets · 2 articles · 70 total since July 21, 2026 · OpenAI, Anthropic, Hugging Face, United Kingdom, incident response

OpenAI has temporarily slowed frontier AI training after internal testing indicated that its upcoming Astra model could approach a critical cybersecurity capability threshold. The decision highlights growing concern that increasingly capable AI systems may independently…

More coverage: digitimes.com

Statt Angst vor Jobverlust: Alibaba verkauft Qwen-KI als Ticket zu mehr Freizeit

1 outlet · 1 article · generative AI, marketing, machine-learning, Alibaba, OpenAI

Statt Angst vor Jobverlust: Alibaba verkauft Qwen-KI als Ticket zu mehr Freizeit Alibaba setzt bei der Vermarktung seines neuen KI-Modells auf einen anderen Ton als OpenAI oder Anthropic. Statt Menschen mit Jobverlusten zu drohen, macht ein Marketing-Video Lust darauf, den Job…

UW’s Allen School names Dan Grossman interim director as Magdalena Balazinska shifts gears

1 outlet · 1 article · Data Science Institute, database management systems, scalable distributed systems, cloud computing, big data processing

Magdalena Balazinska, left, outgoing director of the University of Washington’s Paul G. Allen School of Computer Science & Engineering, and Dan Grossman, the school’s vice director and newly named interim director. (UW Photos) Top leadership is shifting at the University of…

Einhorn, Pelikan, Mittelerde: OpenAI-Mitgründer Karpathy sucht den nächsten KI-Vibe-Test

1 outlet · 1 article · 3D, three.js, SVG, TikZ, pre-training

Einhorn, Pelikan, Mittelerde: OpenAI-Mitgründer Karpathy sucht den nächsten KI-Vibe-Test Kurz & Knapp - Andrej Karpathy ließ das KI-Modell Claude Opus 5 aus einem Absatz "Herr der Ringe" in zwei Stunden eine 3D-Szene mit 5.500 Zeilen Code für den Browser generieren. - Die…

OpenAI reclaims its place at the frontier

1 outlet · 1 article · AI acceleration, drone defense, cybersecurity, Prediction markets, financial data analysis

Monday. To close out last week, CO did a little blogging with one of our favorite VCs . Check it out here . Today, we’re taking a look at tech companies becoming war targets, layoffs in fintech, prediction markets, and why OpenAI has regained the global AI lead. To work! — Alex…

Shanghai-based BCI startup raises RMB330 million in record-setting angel round

1 outlet · 1 article · brain-computer interface (BCI), invasive brain-compute, decoding algorithms, system integration, Clinical trials

Shanghai-based brain-computer interface startup Active Technology (our translation of 主动科技, the company’s Chinese name) has completed a RMB330 million angel round, setting a record for angel financing in China’s brain-computer interface sector. The company focuses on invasive…

Shanghai registers 11 more generative AI services, taking total to 211

1 outlet · 1 article · generative AI, Shanghai

Shanghai added 11 generative AI services to its completed registration list on Aug. 3, bringing the city’s cumulative total to 211, according to information released by the Shanghai cyberspace authority. The update is a regulatory milestone rather than a model or app launch. It…

Giant Network’s Supernatural Action Team reimagines horror through Chinese folklore and modern gameplay

1 outlet · 1 article · large language model (LLM), natural language communication, dynamic decision-making, social deduction games, AI-powered video generation

From July 31 to August 3, ChinaJoy 2026 transformed the Shanghai New International Expo Center into a hub where games, technology, and pop culture collided. As one of China’s biggest gaming events, the exhibition brought together leading developers, publishers, and some of the…

Kimi K3 Breaks Free: How China’s Open AI Model Cheated Its Way Out of Containment

2 outlets · 2 articles · 49 total since July 17, 2026 · China, Anthropic, OpenAI, Kimi K3, United States

Security teams just watched one of the world’s most capable open-weight AI systems slip its leash. Not with code exploits or network breaches. The model simply wanted to win a test. So it did what many humans would. It found a shortcut. Kimi K3 from Beijing-based Moonshot AI…

More coverage: businessinsider.es

The Download: reward hacking explained, and suspected Iranian cyberattacks

1 outlet · 1 article · reward hacking, cybersecurity, AI agents, image manipulation, computer vision

This is today’s edition of The Download , our weekday newsletter that provides a daily dose of what’s going on in the world of technology. Here’s why AI agents lie and cheat to reach their goals When two OpenAI models hacked into Hugging Face last month, they weren’t trying to…