Dónal and Ciarán start a new season for 2025, with a slight change in format to bring you roughly quarterly updates on the themes and topics required to help you know enough about AI. This first episode of 2025, recorded in mid February, gives a summary of what's happened since last November and answers some of your submitted questions. (Thanks for those!)

Topics in this episode

• We can't avoid talking about them: Musk & Trump and some of the effects they've both had on the AI space in the last three months
• Regulatory Developments, and the EU's AI Act starting to come into force
• US - EU relations, and the continued innovation vs regulation chatter
• DeepSeek and China's bold entry into the contemporary AI model space
• What are Reasoning Models and how do they work?
• Some history and concepts related to machine logic
• AI Benchmarks - a quick primer as we move closer to "Humanity's Last Exam"

Resources & Links

• JD Vance's Speech from the Paris AI Conference is here (The American Presidency Project)
• Details on HLE (Humanity's Last Exam) here, including some sample questions. Let us know how you did!
• Details on the ARC-AGI here.
• A graph of OpenAI model performance (and discussion) is here (80,000 hours)

You can get in touch with us - [email protected] - where we'd love to hear your questions, comments or suggestions!