The Inktelligence - June 30, 2005

Want to catch up on the latest state of generative AI? Here’s a massive report by Innovation Endeavors that’s worth a read. Be forewarned that it’s a 126-page deck so grab a cup of coffee and take your time to go through it.

TL;DR:

• Generative AI has gone mainstream – 1 in 8 workers worldwide now uses AI every month, with 90% of that growth happening in just the last 6 months.

• AI-native applications are now well into the billions of annual run rate. Case in point - Cursor, the AI-coding application, is the fastest SaaS to reach 100M ARR. Scaling continues across all dimensions – All technical metrics for models continue to improve >10x year-over-year, including cost, intelligence, context windows, and more. The average duration of human task a model can reliably do is doubling every 7 months.

• The economics of foundation models are…confusing – OpenAI & Anthropic are showing truly unprecedented growth, accelerating at $B+ of annual revenue. But, end-to-end training costs for frontier models near $500M, and the typical model become obsolete within 3 weeks of launch thanks to competition & open source convergence.

• Just like the smartest humans, the smartest AI will “thinks before it speaks” – Reasoning models trained to think before responding likely represent a new scaling law — but training them requires significant advances in post-training, including reinforcement learning & reward models. Post-training may become more important than pre-training.

• AI has now infiltrated almost all specialist professions – From engineers and accountants to designers and lawyers, AI copilots and agents are now tackling high-value tasks in virtually all knowledge worker domains.

• Agents finally work, but we are early in understanding how to build AI products – Agents have finally hit the mainstream, but design patterns & system architectures for AI products are still extremely early.

• “AI-native” organizations will look very different – Flatter teams of capable generalists will become the norm as generative AI lessens the value of specialized skills. Many roles will blur - such as product, design, & engineering.

One of the most interesting set of charts I found in the presentation is the one originally published by Google Research from this paper. There seems to be a threshold in scaling before the performance of models takes off. This is why the AI foundation model companies are investing a huge amount of capital up front before they see any profit. Maybe profits for AI foundation model companies will also display “emergent” behavior?