FTC removes Lina Khan-era posts about AI risks and open source (techcrunch.com)

The FTC quietly removed three blog posts from the Lina Khan era that flagged AI risks and the implications of open‑weights foundation models: “On Open‑Weights Foundation Models” (July 10, 2024), “Consumers Are Voicing Concerns About AI” (Oct 2023), and a staff‑authored “AI and the Risk of Consumer Harm” (Jan 3, 2025) that warned the agency was “taking note of AI’s potential for real‑world instances of harm — from incentivizing commercial surveillance to enabling fraud and impersonation to perpetuating illegal discrimination.” TechCrunch reports the deletions follow a broader March purge of roughly 300 FTC posts and come after new Chair Andrew Ferguson’s leadership reshuffle and a shift in federal priorities under the Trump administration. For the AI/ML community this matters because it signals a narrowing of public regulatory guidance on safety, consumer harm, and the risks of widely shared model weights at a moment of rapid model proliferation. Removing historic analysis reduces transparency for researchers, startups and compliance teams who rely on agency signals to design safety mitigations, red teaming and governance processes. The takedowns also raise legal and archival concerns — potential violations of the Federal Records Act and Open Government Data Act — and reflect tension between a federal push toward faster growth and open‑source support and the need for guardrails that address surveillance, fraud, impersonation, and discrimination risks.