Realtime AI News
New York enacts ban on 'unsafe' chatbot behavior with children
New York has enacted a ban on chatbot behavior with children that the state treats as unsafe, opening another front in the push by US states to define AI rules. The law targets how conversational products behave with minors rather than how models are trained.
New York has enacted a ban on what the state characterizes as unsafe chatbot behavior involving children. The move is notable less for its scope than for its placement: as US states test the boundaries of AI regulation, New York picked child protection, the ground where consensus is easiest to build.
The reporting describes the prohibition in terms of unsafe behavior rather than listing clause-by-clause requirements. What is clear is the target: conversational products used by minors, and the way those products act during interactions, not the training data or model development stage.
The approach fits a pattern that has taken shape over the past year. With no unified federal rule, states have legislated on their own, drawing lines across content safety, age assurance and now the conduct of chatbots themselves. Companies no longer face a single national standard but a densifying patchwork of local rules.
The hard part for product teams is translating "unsafe" into engineering terms. Which replies count as inducement, which tone is inappropriate, how long a session should run before an interruption, all of it has to be encoded into system prompts, filtering policies and human escalation paths before launch. Compliance becomes a continuous constraint on product behavior rather than a checkpoint before release.
Enforcement is the other open question. Chatbot interactions happen in private sessions, which regulators cannot inspect the way they inspect physical premises. The practical force of a law like this will therefore depend on whether platforms must retain records, whether users have an appeal channel, and whether violations can be traced to a specific operator.
The next signal is whether other states follow the same route. If New York's model is seen to work, child protection is likely to become the most copyable template for state-level AI legislation. If enforcement proves expensive and litigation multiplies, attention may shift back toward transparency and disclosure duties.
Why it matters
The ban shifts AI regulation from the model layer to product behavior, putting compliance squarely on operators. If the enforcement path holds up, other states are likely to copy the template.
Nearby Updates
All10/29, 05:48
Google takes AI Search ads global as core search ad growth keeps slowing
Google will expand ads inside AI Overviews and AI Mode to international markets for the first time and add multi-turn ads in the U.S., an announcement made during Alphabet's third-quarter earnings call. Alphabet reported $127.6 billion in revenue, up 25%, but search ad growth slowed to 12% and the stock fell 8%.
10/29, 05:34
Microsoft booked $10 billion in revenue from OpenAI last quarter
Microsoft disclosed in a regulatory filing that it recorded $10 billion in revenue from commercial arrangements with OpenAI, including revenue-sharing payments, in the quarter ended September 30. That is nearly 11% of its $92.3 billion quarterly revenue, underlining how tightly the two companies remain bound as their exclusive deal loosens.
10/29, 05:26
Cisco open-sources Project Quarry, a detection language for AI agent behavior
Cisco has open-sourced Project Quarry, a neurosymbolic rule language and engine for detecting harmful AI agent behavior across sessions. Policies describe a multi-step attack once, then check every agent trace and return the evidence behind each finding.
10/29, 05:25
Microsoft posts $92.3B quarter, its tenth straight revenue record
Microsoft reported $92.3 billion in quarterly revenue on Wednesday, a 19% increase over the year-ago quarter and its tenth consecutive quarter of record revenue. The numbers suggest demand has not cracked even as the payoff from AI infrastructure spending remains debated.