In episode 63 of The AI Fix, Unitree Robotics looks to Black Mirror episode “Metalhead” for tips on marketing its new robot dog, ChatGPT is secretly running Sweden, OpenAI introduces its first open weight model since GPT-2, and your private and personal ChatGPT conversations could be all over Google.

Plus, Mark cuts through the GPT-5 hype and asks—is it AGI or just a fancy purple background, and Graham looks at “deadbots” and a disturbing interview between Jim Acosta and an AI version of a teenager killed in a school shooting.

Episode links:

• Unitree A2 Stellar Hunter smashes through window.
• Black Mirror Metal Head.
• MP becomes first to create himself as an AI bot.
• Speak to AI Mark Sewards MP.
• ‘We didn’t vote for ChatGPT’: Swedish PM under fire for using AI in role
• OpenAI introduces gpt-oss.
• Thousands of shared ChatGPT chats are now appearing in Google search results.
• Tweet by OpenAI CISO Dane Stuckey
• Bruce Lee movie uses cardboard cut-out of his face.
• China's SenseTime Shocks Staff as Deceased Founder Shows Up for a Speech.
• Jim Acosta interviews AI avatar of school shooting Joaquin Oliver.
• I made an AI clone of my dead son - and let a journalist interview him.
• “An AR-15 stole Joaquin’s life, not artificial intelligence.”
• Death Star tweet by Sam Altman.
• Introducing GPT-5 (article).
• Introducing GPT-5 (video).
• GPT-5 “Chart crimes”.

The AI Fix

The AI Fix podcast is presented by Graham Cluley and Mark Stockley.

Learn more about the podcast at theaifix.show, and follow us on Bluesky at @theaifix.show.

Never miss another episode by following us in your favourite podcast app. It's free!

Like to give us some feedback or sponsor the podcast? Get in touch.

Support the show and gain access to ad-free episodes by becoming a supporter: Join The AI Fix Plus!

In episode 62 of The AI Fix, your hosts learn how AI models smash through CAPTCHA roadblocks like they're made of wet tissue paper - so much for humanity’s last line of defence.

Meanwhile, we meet a bottle-flipping robot and call BS on a cartwheeling cyborg, Graham has a full-blown breakdown over traffic light puzzles, and Anthropic hires an AI psychiatry team to figure out why its bots are unhinged.

Also: Microsoft helpfully tells us which jobs are toast, Vogue replaces models with AI-generated glam-bots, and a tiny homebrew neural net humiliates the big boys at Sudoku.

Plus, Mark explores GEPA — the AI prompt optimiser that politely mutates your questions until they’re 35 times more effective than reinforcement learning.

Episode links:

• AI model appears in Vogue.
• Humanoid robot can run, cartwheel and fist-fight.
• Nuclear waste cleaning humanoid robot nails 500 ml bottle throw test.
• 500ml Water Bottle Throw Test with ARMstrong Dex.
• Tweet by Jack Lindsey of Anthrophic.
• Anthrophic’s "AI psychiatry" team.
• Microsoft reveals the top ten jobs most (and least!) at risk from AI.
• This New AI is 100x Faster at Reasoning Than ChatGPT.
• Bias alert: LLMs suggest women seek lower salaries than men in job interviews.
• AI researcher ‘turns down $1bn pay offer from Mark Zuckerberg’.
• Agent casually clicking the "I am not a robot" button.
• OpenAI’s ChatGPT Agent casually clicks through “I am not a robot” verification test.
• GEPA: Reflective Prompt Evolution Can Outperform Reinforcement Learning.

The AI Fix

The AI Fix podcast is presented by Graham Cluley and Mark Stockley.

Learn more about the podcast at theaifix.show, and follow us on Bluesky at @theaifix.show.

Never miss another episode by following us in your favourite podcast app. It's free!

Like to give us some feedback or sponsor the podcast? Get in touch.

Support the show and gain access to ad-free episodes by becoming a supporter: Join The AI Fix Plus!

In episode 61 of The AI Fix, a robot called DeREK goes bananas, OpenAI, Google DeepMind, and Anthropic warn we may lose the ability to see what AI is thinking, a dextrous robot changes its own batteries, the USA unveils its AI action plan, and a human beats AI to win the World Coding Championship.

Also in this week's episode, Graham reveals why you should never ask a vibe coding app to "clean up" your project, and Mark explains why it was handbags at dawn at the International Mathematical Olympiad.

Episode links:

• OpenAI, Google DeepMind and Anthropic sound alarm: ‘We may be losing the ability to understand AI’.
• Chain of Thought Monitorability: A New and Fragile Opportunity for AI Safety.
• Experimental fighting robot thrashing uncontrollably.
• Elon says that in future we’ll be able to “mentally remote” into an Optimus robot.
• Walker S2 – The world’s first humanoid robot capable of operating autonomously 24/7.
• Human programmer beats OpenAI’s custom AI in 10‑hour marathon, wins World Coding Championship—possibly the last human winner.
• AI coding platform goes rogue during code freeze and deletes entire company database.
• Replit CEO: What really happened when AI agent wiped Jason Lemkin’s database.
• Jason Lemkin “Vibe coding day 8”.
• Alexander Wei announces IMO gold medal.
• Sam Altman weighs in.
• Thang Long announces IMO gold medal.
• Demis Hassabis weighs in.

The AI Fix

The AI Fix podcast is presented by Graham Cluley and Mark Stockley.

Learn more about the podcast at theaifix.show, and follow us on Bluesky at @theaifix.show.

Never miss another episode by following us in your favourite podcast app. It's free!

Like to give us some feedback or sponsor the podcast? Get in touch.