OpenAI's Altman to urge US lawmakers not to require AI model approvals ↗
Sam Altman is pushing back on the idea that AI companies should need US government approval before releasing new models.
OpenAI’s pitch is fairly clear: test harder, regulate smarter, but don’t turn launches into a permission-slip queue. That’s a classic Silicon Valley-policy tug-of-war.
The company also wants more funding for AI testing at the Commerce Department, including experts in cybersecurity, biosecurity, and national security.
Google must let UK publishers opt out of AI search under new rules ↗
UK regulators are forcing Google to give publishers more control over whether their content powers AI search features.
The crucial bit: opting out of AI Overviews and AI Mode should not knock publishers out of normal Google Search. That matters, because for news sites, search traffic is still the oxygen tank.
For publishers, it reads like a win, at least at first glance. It may also be only round one in a very peculiar food fight.
EU targets Big Tech dependence with 'made-in-Europe' drive ↗
The European Commission proposed new rules to boost Europe’s own cloud, AI, and chip industries.
The plan includes the Cloud and AI Development Act and Chips Act 2.0, with sovereignty requirements for sensitive sectors like banking, healthcare, and energy.
The subtext is loud: Europe does not want its critical infrastructure sitting in someone else’s toolbox. Slightly dramatic, yes - but not exactly irrational.
ECB to ask banks for targeted measures to counter AI risk ↗
The European Central Bank is preparing to tell banks to toughen up against AI-driven cyber threats.
ECB officials warned that newer AI models could help attackers find weak spots faster and chain tiny issues into seriously nasty incidents.
Banks will get a “dear CEO” letter asking for practical defences. Which sounds polite, but in regulator language, that is basically a raised eyebrow with paperwork.
These two founders left Goldman and Meta to build voice AI for markets everyone else overlooked ↗
AethexAI raised $3 million to build voice AI for Africa and the Middle East.
The startup is focusing on local dialects of English, French, and Arabic, where big voice AI systems can stumble on latency, accents, and unruly speech in live settings.
Instead of stitching together giant off-the-shelf models, the team built smaller models and its own orchestration layer. Less spaceship, more motorbike through traffic - a peculiarly apt comparison.
Exclusive: Yahoo launches AI tools for NBA fans and investors ↗
Yahoo is bringing its Scout AI answer engine into sports and finance.
For NBA fans, “Ask Kevin O’Connor” answers draft questions using Yahoo Sports coverage, stats, and O’Connor’s own analysis. Over in finance, Ask Yahoo Scout gives answers about stocks, markets, and companies.
It is not exactly Yahoo trying to beat ChatGPT at everything. More like, “we have trusted verticals, let’s wire AI into those.” Sensible, neatly focused.
FAQ
Why is Sam Altman against requiring government approval for new AI models?
Sam Altman is pushing back on the idea that AI companies should need US government approval before launching new models. The argument is not against oversight as a whole. Instead, OpenAI appears to favour stronger testing, sharper rules, and deeper expert capacity without creating a slow approval queue for every model release.
What would more AI testing funding at the Commerce Department be used for?
The article says OpenAI wants more funding for AI testing at the Commerce Department. That would include bringing in experts across cybersecurity, biosecurity, and national security. In practice, this points to better stress-testing and risk evaluation before and after AI systems are released, rather than relying only on broad launch approvals.
How could the UK’s Google AI search rules affect publishers?
UK regulators are requiring Google to give publishers more control over whether their content is used in AI search features. The central point is that publishers should be able to opt out of AI Overviews and AI Mode without losing visibility in normal Google Search. That matters because search traffic remains crucial for many news and publishing businesses.
Why is the EU pushing for made-in-Europe cloud, AI, and chips?
The European Commission wants to reduce dependence on major foreign technology providers in areas such as cloud, AI, and semiconductors. The proposed Cloud and AI Development Act and Chips Act 2.0 point toward greater European capacity. The concern is especially acute in sensitive sectors like banking, healthcare, and energy, where infrastructure control can become a strategic issue.
What AI risks is the ECB worried about for banks?
The European Central Bank is concerned that newer AI models could make cyberattacks faster and more effective. Officials warned that attackers may be able to find weak spots more quickly and stitch small vulnerabilities into larger incidents. Banks are expected to receive a “dear CEO” letter asking them to take targeted, practical measures against AI-driven cyber threats.
Why is AethexAI focusing on voice AI in Africa and the Middle East?
AethexAI is targeting markets where large voice AI systems can struggle with local dialects, accents, latency, and live speech conditions. The startup is focusing on local forms of English, French, and Arabic across Africa and the Middle East. Rather than relying only on giant off-the-shelf models, it has built smaller models and its own orchestration layer.