AI's opportunity cost, in frontier research.
From Hierarchy to Intelligence.
Jack Dorsey and Roelof Botha on Block's flat-org redesign after roughly 4,000 of 10,000 roles were retired. The most-cited proof point for service-as-software inside a public company.
Reflexive AI usage is now a baseline expectation.
Tobi Lütke's confirmed internal memo declares AI usage non-optional, adds AI questions to performance and peer reviews, and requires teams to prove the work cannot be done with AI before any new headcount is approved.
AI assistant handled 700 agents' worth of work in month one.
2.3M conversations, resolution time from 11 minutes to under 2, projected $40M of 2024 profit improvement. The operator press release every European AI-restructure deck cites first.
Multi-agentic underwriting with Palantir and Anthropic.
Publicly-traded carrier disclosing an agent-based underwriting and claim-review stack at investor day. Not a pilot, the production stack for one of the largest insurance balance sheets in the world.
AI agents for industrial automation.
Siemens introduces AI agents inside its industrial automation stack, with deployments across manufacturing and process industries rather than positioning AI as a roadmap item.
AI startups: sell work, not software.
Tavel reframes the SaaS playbook. Instead of selling per-seat productivity, AI-native firms should sell the completed work itself, replacing the cost of labor rather than discounting human time by 15%.
Generative AI's Act Two.
Sonya Huang and Pat Grady argue Gen AI is transitioning from technology-out Act One to customer-back Act Two, where enduring value comes from solving end-to-end human problems rather than novelty chat.
Generative AI at Work.
Field study of 5,179 customer-support agents using a GenAI chat assistant found a 14% average productivity boost, with a 34% gain for novices and minimal effect on experienced workers. Attrition fell 8.6%, sentiment rose.
Machines of Loving Grace.
Anthropic's CEO sketches a compressed 21st century in which powerful AI accelerates discovery 10 to 100 times across biology, neuroscience, economic development, and work, achieving roughly a century of progress in 5 to 10 years.
Send anything we should add at hello@datajockey.us.