In episode 46 of The AI Fix, China trolls US tariffs, a microscopic pogoing flea-bot makes a tiny leap forward for robotics, Google unveils the Agent2Agent protocol, a robot dog is so cute it ruins Graham's entire day, and Europe commits €20 billion and all of its buzzwords to five moonshot AI gigafactories.

Graham brings his clone to work for a game of “Cluley or Cloney”, and Mark introduces Graham to a mind-reading technology that could transform the lives of people with serious neurological disorders.

Episode links:

• Loona - the intelligent Petbot.
• Robots have learnt to hop to a height of 20 cm.
• Tiny hopping robots.
• Google announces A2A, the Agent2Agent Protocol.
• Donald Trump’s education secretary talks about “A1”.
• Linda McMahon speaks at 2025 ASU+GSV Summit.
• China mocks Trump’s tariffs with AI video of US sweatshop workers: ‘Make America Great Again’.
• AI murder predictor catches killers before they strike.
• UK: Ministry of Justice secretly developing ‘murder prediction’ system.
• EU to build AI gigafactories in €20bn push to catch up with US and China.
• ElevenLabs.
• Magic genies, killer robots and AI friends – all the best quotes from Elon Musk's chat with Rishi Sunak in London
• PauseAI.
• A streaming brain-to-voice neuroprosthesis to restore naturalistic communication.
• Video of streaming brain-to-voice communication.

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

In episode 45 of The AI Fix, our hosts discover that ChatGPT is running the world, Mark learns that mattress companies have scientists, Gen Z has nightmares about AI, OpenAI gets a bag, Graham eats too many cheese sandwiches, and too much training makes AIs over-sensitive.

Mark reveals why he's got beef with cows, GPT-4.5 beats the Turing test, and Anthropic’s brain scanner reveals how AIs really think.

Episode links:

• OpenAI Secures $40 Billion to Pursue Next-Gen AI.
• Lots of people are having nightmares… about AI.
• Researchers warn of ‘catastrophic overtraining’ in LLMs.
• Hackers can probe your robot dog’s backdoor.
• White House appears to use AI in tariff rate calculations.
• Can you tell a human from an AI?
• Tracing the thoughts of a large language model.

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

In episode 44 of The AI Fix, ChatGPT won’t build a crystal meth lab, GPT-4o improves the show’s podcast art, some students manage to screw in a lightbulb, Google releases Gemini 2.5 Pro Experimental and nobody notices, and Mark invents a clock for measuring AI time.

Graham explains how ChatGPT’s love for Young Adult fiction can be used to turn it into an evil malware developer, and Mark looks at the ARC-AGI-2 benchmark and a staggering leap forward in AI intelligence.

Episode links:

• Engineers develop hybrid robot that balances strength and flexibility - and can screw in a lightbulb.
• Gemini 2.5.
• Introducing 4o Image Generation.
• New LLM Jailbreak Technique Enables Easy Creation of Password-Stealing Malware.
• New AI Attack Compromises Google Chrome’s Password Manager.
• The ARC Prize.
• The ARC test with which Mark challenged Graham.

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

In episode 43 of The AI Fix, our hosts discover a robot that isn't terrifying, a newspaper shuns journalists in favour of AI, Graham watches a robot dog learn to stand, an AI computer programmer develops a familiar attitude, and New York tries to stop its humans arming their robots.

Graham worries about AI vomit, a Norwegian man is falsely accused of murder by ChatGPT, and Mark looks at why we're suddenly surrounded by news about robots.

Episode links:

• Introducing IntuiCell.
• New York law would ban civilians from weaponizing robots.
• Jensen Huang Introduces Blue: NVIDIA & Disney Research’s AI Robot.
• AI coding assistant refuses to write code, tells user to learn programming instead.
• AI can forecast the weather in seconds without needing supercomputers.
• Italian Newspaper Creates Entirely AI-Generated Edition.
• AI hallucinations: ChatGPT created a fake child murderer.
• Exclusive: Russian disinformation floods AI chatbots, study finds.
• Walk, Run, Crawl, RL Fun.
• Elon Musk: 10 billion humanoid robots by 2040 at $20K-$25K each.
• Elon Musk tells Ted Cruz he puts the odds of “killer robots annihilating humanity” at 10-20%.

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

In episode 42 of the AI Fix, our hosts discover why ads for the Neo Gamma robot are so sinister, Graham plays peek-a-boo with a crow, humans give up writing, an AI designs a drug, an upstart AI agent gets everyone’s attention, and a talking fish offers our hosts some sage advice.

Graham wonders if AIs have feelings, and Mark introduces Graham to the reversal curse and explains why AIs don’t know what happened.

Episode links:

• When AI met Billy Bass fish.
• Manus.
• AM Turing award.
• NEO Gamma robot.
• Insilico Medicine's AI-driven drug Rentosertib receives official generic name.
• The Widespread Adoption of Large Language Model-Assisted Writing Across Society.
• Assessing and alleviating state anxiety in large language models.
• ChatGPT on the Couch? How to Calm a Stressed-Out AI.
• Geoffrey Hinton on AI having feelings.
• Intuitive physics understanding emerges from self-supervised pretraining on natural videos.
• Episodic Memories Generation and Evaluation Benchmark for Large Language Models.
• The Reversal Curse: LLMs trained on "A is B" fail to learn "B is A".

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

In episode 41 of the AI Fix, our hosts learn that society needs to be completely reordered by December, Grok accuses Trump of being a Russian asset, Graham discovers that parents were wrong about computer games all along, and Mark wonders if a kung-fu kicking robot from Unitree is the hero that we need.

Graham gives an AI a Rorschach test and learns about "Norman" the psychopathic AI, and Mark discovers why we should actually be optimistic about AI.

Plus - don’t miss our featured interview with Marc Beckman, the author of “Some future day: How AI is going to change everything.”

Episode links:

• Amazon is reportedly developing its own AI ‘reasoning’ model.
• The kung-fu kicking Unitree G1 robot.
• Elon Musk’s AI, Grok, thinks Donald Trump is a Russian asset.
• People are using Mario to benchmark AI.
• Anthropic CEO says AI could be smarter than all humans by next year.
• A factory in Shanghai has made a robot army.
• We gave an AI a Rorschach test. What it saw in the inkblots offers a window into the human mind.
• Project Norman.
• Some Future Day: How AI Is Going to Change Everything.

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

In episode 40 of the AI Fix, Graham meets a shape-shifting GOAT, a robot dog gets wet, Mark likes Claude 3.7 Sonnet, OpenAI releases its dullest model yet, Grok 3 needs to go home and have a lie down, and everyone loses their minds over two AI agents booking a hotel room using 90s-era modem dial-up sounds.

Graham tells the incredible story of a woman whose life was saved after ChatGPT told her to go to the emergency room, stat! And Mark explains how just a little negativity made GPT-4o bad to the bone.

Episode links:

• Underwater Walking of the Honey Badger Legged Robot.
• Claude 3.7 Sonnet and Claude Code.
• Morphing robot GOAT adapts to diverse terrains.
• Morphing robot turns challenging terrain to its advantage.
• Two AI agents on a phone call realise they’re both AI.
• Gibberlink.
• Grok’s new “unhinged” voice mode can curse and scream, simulate phone sex.
• Introducing GPT-4.5.
• ChatGPT Saved My Life (No, Seriously, I’m Writing this from the ER).
• Emergent misalignment.

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

In episode 39 of the AI Fix, our hosts watch a drone and a robot dog shoot fireworks at each other, xAI launches Grok 3, Mark explains that AIs can design genomes now, a robot starts a punch up, Zuck becomes a mind reader, an AI cracks a ten-year science question in two days, and an anatomically accurate synthetic human recreates a terrifying scene from The Long Good Friday.

Graham learns that it always pays to be polite before running over 15 people with a train, and Mark discovers why AIs value some lives more than others, particularly their own.

Episode links:

• AI can now design the genetic code for life.
• Mark Zuckerberg wants to inject himself into your brain.
• AI cracks superbug problem in two days, ten years quicker than humans.
• Robot dog chained-up… for art.
• Protoclone, the world's first bipedal, musculoskeletal android.
• Dancing Unitree G1 robot.
• Things get a bit lively at robot dance party in front of one billion people.
• xAI releases Grok3.
• Drone and robot dog do battle in dog fight.
• Why you shouldn’t say 'please' or 'thank you' to AI (and why it matters).
• Does it pay to be polite to chatbots?
• Should We Respect LLMs? A Cross-Lingual Study on the Influence of Prompt Politeness on LLM Performance.
• Utility Engineering: Analyzing and Controlling Emergent Value Systems in AIs.

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