... government of the people, by the people, for the people ... — Abraham Lincoln, Gettysburg Address (1863) The cost of AI is dropping rapidly. GPT-4-class capabilities cost roughly $30 per million tokens in early 2023; today the same runs under $1 , and some providers are pushing costs below $0.10 . Across benchmarks, inference prices have fallen between 9x and 900x per year , with a median decline near 50x. Even frontier models are getting dramatically cheaper each generation, with open-source models following closely behind. And crucially, even if “Nobel-Prize-winning genius-level” intelli
The accelerating decline in AI inference costs, highlighted by specific figures, indicates a critical inflection point where advanced AI capabilities become widely accessible.
This dramatic cost reduction democratizes access to powerful AI, enabling pervasive integration into various systems and drastically altering economic and technological landscapes.
The financial barrier to deploying and scaling sophisticated AI models has dramatically lowered, shifting the focus from model access to data infrastructure and agentic systems.
- · AI developers
- · Small businesses
- · AI-powered services
- · Data infrastructure providers
- · Legacy software companies
- · Human white-collar workers (routine tasks)
- · AI providers with high-cost models
Widespread deployment of AI agents across various industries due to affordability.
Compression of many white-collar workflows and SaaS layers, leading to job displacement and new business models.
Emergence of new, highly autonomous enterprises driven almost entirely by AI agents, redefining corporate structures and national productivity.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at Berkeley BAIR Blog