The Register Kettle

This week on The Kettle, host Brandon Vigliarolo is joined by systems editor Tobias Mann and senior reporter Tom Claburn to discuss the current state of locally-installed coding assistant LLMs. After some experimentation, both Claburn and Mann have concluded they're actually getting pretty good.What does that mean for the computing power crunch? Hopefully it means cash-strapped developers will be able to get off the train of increased prices from AI vendors and actually run a useful local code

Windows is a mess, Copilot's been a disaster, updates have been failing ... hey, Microsoft - everything okay?There's been a lot going wrong for the Windows maker of late, and leadership is promising improvements. Join host and El Reg Brandon Vigliarolo, US editor Avram Piltch, and Microsoft reporter Richard Speed to pick apart the situation and read the entrails to see what Microsoft's future holds.

This week on the Kettle we split the conversation between two topics: The biggest news out of Google Cloud Next, and the latest hype-bursting revelations about Anthropic's Mythos AI, and the fact that it's already been accessed inappropriately.Different topics, yes, but both point to the same thing: AI is reshaping everything from Google's hardware designs to the security world, and we'd better get ready: The tech world isn't slowing down for anyone.

Prompt injection attacks are, in a lot of ways, like phishing a human: By convincing them of your harmless intent, they do something that's not in their best nature just like the gullible person handing over corporate details. Thinking about it that way, our current AI paradigm will likely never escape prompt injection attacks - you can't patch out the ability to con someone with language, human or not, it seems. Join host Brandon Vigliarolo, El Reg cybersecurity editor Jessica Lyons, an senior

Anthropic stunned the infosec world this week with the announcement of a new zero-day discovering and exploiting AI called Mythos, which it claims is too dangerous to release to anyone but the biggest companies in the tech world.Join host Brandon Vigliarolo, APAC editor Simon Sharwood, and senior reporter Tom Claburn to discuss the Mythos news, what it means, and why it's probably appropriate to be a bit skeptical of all the company's claims.

A human mistake at Anthropic last week led to the entirety of Claude Code's source code being leaked to the web, pubished on GitHub, forked, dissected, picked apart, and otherwise exposed to the world to see - perfect fodder for a Kettle of vultures looking for some news to pick apart.Join cybersecurity editor Jessica Lyons, senior reporter Thomas Claburn, and host Brandon Vigliarolo as they pick apart the leak of Claude Code's source, its security implications, what's been found in the code, and

Did you know that if you tell your AI it's an expert coder it'll actually write worse code? Join Brandon, systems editor Tobias Mann, and senior reporter Thomas Claburn to discuss the state of - and limitations with - AI software development, and why we're not convinced it's going to replace a human dev anytime soon.

We regret to inform you that RSAC 2026 is probably going to be all about agentic AI, and with good reason: It's changing both the defensive and offensive security paradigms. Along with agentic AI, The Register security editor Jessica Lyons tells host Brandon Vigliarolo that other hot topics at the industry's big event this year are likely to include the war in Iran, North Korean IT employment scammers, and maybe a bit of politics too: The US government won't be represented at all at this year's