6 articles.
Google Gemini keeps conversations 72 hours when activity tracking is off and 18 months by default, with human reviewers able to see them for up to three years. Anthropic keeps consumer chats 30 days on the backend by default and lets you opt in to model improvement. OpenAI lets any consumer opt out of training while business and API data is not used for training by default. Two of the three describe training as opt-in and one describes it as opt-out.
The California AI Transparency Act has been operative since 1 January 2026 and requires covered providers to publish a free AI detection tool and embed both visible and machine-readable provenance in generated images, video and audio. The penalty is $5,000 per violation with each day counted separately. The definition of generative AI includes text, but the operative sections require disclosures only for image, video and audio, so an AI-written blog post carries no disclosure duty at all.
NIST's AI RMF is free, voluntary, and structured around four functions. It gives businesses a concrete way to start governing AI without building a framework from scratch.
Employees pasting sensitive material into unsanctioned AI tools creates data-exposure and compliance risk that most organisations have not governed. The core problem is unapproved tools plus confidential input, not the model itself.
Local Law 144 obliges employers to run an independent bias audit on automated hiring tools, publish it, and notify candidates. It says nothing about fixing what the audit finds. A 2025 FAccT paper analysed all 116 publicly available audits and found most do not report results suggesting a violation of the four-fifths rule, while also showing the underlying tools could often be in violation once missing demographic data is accounted for. HireVue's own blog says whether its tools qualify as AEDTs is the employer's call. Here is where the enforcement pressure is heading.
Vapi's own worked example puts a fully loaded AI phone agent at $0.08 to $0.13 a minute, while Smith.ai sells human agents at $7 to $10 per call. The bigger story is that the FCC rule everyone quotes governs outbound robocalls, not the inbound calls a small business wants to automate, and that the California chatbot law passed on September 28, 2026 exempts every business under $500 million in revenue.