Has AGI Progress Actually Stalled?
Even though ChatGPT and Gemini are getting better, they still suffer from the same types of failure modes, like hallucinations, which suggests that simply scaling them up yet further will not fix these issues. But at the same time we can see around us the adoption and use of ever more AI tools for doing specific types of work as, despite these flaws, they can still be very useful in specific contexts. This video explores this tension between a slowing down of progress towards artificial general intelligence, AGI, and yet a growing impact of AI tools on our political economy. ▬▬ Chapters ▬▬ 00:00 - AGI plateau / AI everywhere 01:02 - Same failure modes remain 07:50 - AGI needs more research 09:58 - Visualising the tension 12:42 - Existing AI tools are useful 14:35 - Impacts on political economy 16:15 - Nano Banana Pro is impressive 18:37 - Getting ahead of the disruption ▬▬ Animated Nano Banana Pro character ▬▬ I created the initial character with Nano Banana Pro and then animated it with Veo. Although, of course, it was far from a one shot process and the final animation required additional editing in DaVinci Resolve. ▬▬ Books referenced ▬▬ The Monetary Theory of Production, Augusto Graziani, Cambridge University Press, 2003 ▬▬ Videos clip quoted ▬▬ AI Explained - Gemini 3 Pro: Breakdown - 19th Nov 2025 • Gemini 3 Pro: Breakdown Andrej Karpathy, co-founder OpenAI, on Dwarkesh Patel - Oct 2025 • Andrej Karpathy — “We’re summoning ghosts,... Demis Hassabis, CEO Google DeepMind, on Hard Fork - Nov 2025 • Demis Hassabis & Josh Woodward Tell Us Why... Llion Jones, co-inventor of the Transformer, on Machine Learning Street Talk (MLST) - Nov 2025 • He Co-Invented the Transformer. Now: Conti... Ilya Sutskever, co-founder of OpenAI, on Dwarkesh Patel - Nov 2025 • Ilya Sutskever – We're moving from the age... Dario Amodei, CEO of Anthropic, at the DealBook Summit - Nov 2025 • Anthropic C.E.O.: Massive A.I. Spending Co... Demis Hassabis on Google DeepMind podcast with Hannah Fry - Dec 2025 • The future of intelligence | Demis Hassabi... Rick Beato - 28 Nov 2025 • The AI Music Race is Over Waymo (Alphabet, USA) - Deploying to London 2026 • Riding with Waymo | Hailing your trip Apollo Go (Baidu, China) - Deploying to Dubai in 2026 • Hello World: Apollo Go Going Global ▬▬ Music ▬▬ Song: Valence - Infinite [NCS Release] Music provided by NoCopyrightSounds Free Download/Stream: http://ncs.io/infinite Watch: • Valence - Infinite | Future Bass | NCS - C...

Something is jamming GPS over Europe. Here's what we found

AI Whistleblower: We Are Being Gaslit By AI Companies, They’re Hiding The Truth! - Karen Hao

The Rise of Sam Altman and OpenAI

Why do tech bros keep misreading this cost crisis chart?

ChatGPT's limits and AI's potential | Michael Wooldridge

Is Your Job in Danger From AI? Gary Marcus Reveals the Truth

Anthopic, OpenAI Should Not Be Allowed to IPO, Says Ed Zitron

Why Chinese AI Is Suddenly So Good (ft. DeepSeek, SeeDance 2.0) | AB Explained

China’s Next AI Shock Is Hardware

We're Entering The Most Dangerous Phase Of AI Yet | AI Architects

They Lied to You About AI (This Study Proves It)

Anthropic Vs. OpenAI: How Safety Became The Advantage In AI

Why full, human level AGI won't happen anytime soon

Yann LeCun's $1B Bet Against LLMs

Opus 4.8 Drops, Demis Hassabis Predicts AGI, and the $220B Foundation | EP #260

Demis Hassabis On What AI Will Do Next

The Singularity Countdown: AGI by 2029, Humans Merge with AI, Intelligence 1000x | Ray Kurzweil

Will AI outsmart human intelligence? - with 'Godfather of AI' Geoffrey Hinton

Yann LeCun: We Won't Reach AGI By Scaling Up LLMS

