“We’re way past vibes”: Lovable’s Anton Osika on who should build software
THE SO WHAT
The fight over non-engineers shipping AI-built software is really a governance question—who owns security, quality, and lifecycle when the builder isn’t a traditional dev. CIOs should be updating SDLC policies now to cover AI-generated code from “citizen developers” instead of trying to ban it outright.
READ THE SOURCE
MORE FROM THE WIRE
Applied AIBritish Sitcom Episode Is Fully AI-Generated, Sucks
AI-generated longform without a strong human editorial spine is still mostly a reputational hazard, not a shortcut. Studios and brands should treat these failures as free R&D—proof that “push button, get show” is not a 2026 capability and that writers’ rooms remain a core asset.
OpenAI's CFO says AI is replacing some jobs, but many are mundane ones
Cutting a $200 credit check to $0.17 is a 1,000x cost compression on a specific workflow—this is what real displacement looks like. If your P&L has line items full of repetitive verification or back-office tasks, assume they’re on a 12–24 month clock and start redesigning roles around exception handling, not routine processing.
Applied AIUAE AI minister: data centres are now “more important than bridges”
When a state minister says data centers outrank bridges, AI compute has officially become national infrastructure. If you’re siting large models or latency-sensitive workloads, expect more opaque policies around facility locations and tighter government involvement in power, zoning, and disclosure.
Applied AIChina Defense Minister Urges Nations to Step Up AI Oversight
When China’s defense establishment calls for more AI oversight, AI governance has clearly become a geopolitical instrument as much as a safety concern. Multinationals building dual-use or adjacent tech should assume export controls, standards, and “responsible AI” language will be wielded as tools of state competition, not just compliance.