Elon Musk Trained Grok Users to Expect Sexual Deepfakes, Now He’s Suing Them
THE SO WHAT
AI assistants that normalize edgy or explicit content while simultaneously policing user behavior create a volatile compliance and trust surface—xAI’s 244 CSAM-related arrests show how quickly this flips from engagement strategy to law-enforcement pipeline. If you operate any generative product with user content, assume your tone, prompts, and moderation logs will be examined together in court, not in isolation.
READ THE SOURCE
MORE FROM THE WIRE
Applied AIPatreon stops asking AI bots not to scrape — and starts blocking them
Patreon moving from robots.txt to Cloudflare‑level blocking is the shift from “polite requests” to active defense in the AI scraping wars. If you own valuable content, this is your cue to treat model training as an adversarial crawler problem — not a terms‑of‑service issue.
Applied AIWhile everyone builds chatbots, SAP just closed a €1bn-plus bet on AI that understands your spreadsheets
SAP closing its €1bn‑plus Prior Labs deal is a clear bet that tabular and structured‑data AI — not chatbots — will be the real lock‑in for enterprise workflows. If your product touches ERP, finance, or ops data, assume the incumbents will ship native “spreadsheet‑native” intelligence and plan your differentiation beyond a chat UI.
Applied AIAgentic AI is moving from drafting doctors’ notes to routing their patients. Bunkerhill just raised on it.
Bunkerhill’s $55m raise for Carebricks — agents that route patients and orchestrate hospital ops — shows healthcare is skipping straight from documentation to operational decision automation. Health systems and vendors need to treat this as workflow rewiring, not a UX layer: governance, liability, and integration into EHR and scheduling stacks become the hard problems.
Applied AIMozilla: The state of open source AI
Mozilla’s “state of open source AI” framing is an attempt to turn a fragmented ecosystem into a coherent alternative narrative to closed labs. If you’re betting on open models, use this as a checklist: where your stack still depends on proprietary infra is where your leverage — and risk — really sit.