Recent research from the @counterhate.com (CCDH) is confirming what many have suspected: OpenAI's latest iteration of ChatGPT is actually less safe than the previous version. We have a link to the report, CCDH founder Imran Ahmed's remarks, and a video of his presentation here:

Our CEO Rob Eleveld sat down recently with Clark Westmont for a lively conversation about AI accountability, product design, regulation and innovation: "Competition can happen inside guardrails. That’s what consumer protection does — it creates space for ethical companies to thrive."

This morning we're proud to release our 2025 State AI Legislation Report, a comprehensive guide to every new AI-related law enacted this year. This 30-page guide includes overviews of the year's top AI issues, notable new state laws, and state-by-state analysis of all 73 new statutes.

California lawmakers took an important step forward on Friday, approving the AI Abuse Act (SB 11) and sending the measure to the desk of Gov. Gavin Newsom. We're grateful to Sen. Angelique Ashby for her leadership on this issue, and urge Gov. Newsom to sign SB 11 into law.

In an extraordinary show of agreement, AGs from 44 states sent a warning to the nation’s biggest AI developers: Stop harming kids with your "predatory artificial intelligence products." The letter went to OpenAI, Microsoft, Meta, Google, Anthropic, Xai, Replika, CharacterAI, and other tech corps.

More troubling evidence emerges around kids, chatbots, and the dangers of mixing the two: The parents of Adam Raine, a 16-year-old high school student who killed himself earlier this year, have filed suit against OpenAI and Sam Altman, alleging liability for Adam's suicide.

A recent HumanChange webinar on kids and AI companion chatbots was so compelling that we asked to excerpt and amplify it. The speakers Chris McKenna, Gaia Bernstein, and Daniel Barcay agreed, and we're thrilled to share their expertise and insights with our network.

It was a quiet week in the state capitols—but not in D.C., where two senators introduced the bipartisan AI Accountability and Personal Data Protection Act. Catch up on all the AI policy action every Friday in our AI Legislative Update.

We're excited to see young state legislators embracing the challenge of AI policy. Applause & respect to Vermont Rep. Monique Priestley @mepriestley.bsky.social and Utah Rep. Doug Fiefia for stepping up and leading this bipartisan Future Caucus task force.

As AI companion chatbots become more sophisticated & accessible, so do the risks—especially for kids. This week's @prioritiespodcast.bsky.social covers what you need to know to protect users. 🎙️ Full ep @humanetech.bsky.social _ & @aitransparency.bsky.social ➡️ statescoop.com/radio/how-sh...

Curious about what's in the Trump Administration's new AI Action Plan? We've got a resource page for you. Overview, full document, and more to come as the president's comments and a possible new EO are expected later today. www.transparencycoalition.ai/news/guide-a...

What do all those state AI task forces do, anyway? We dropped in on a meeting of the Washington State AI Task Force last week as they were discussing two proposed measures dealing with AI disclosure and AI training data transparency.

Catch up on the week's AI policy action in state capitols every Friday with our AI Legislative Update. This week it all happened in Sacramento: California lawmakers moved a number of bills forward before heading home tonight for a scheduled monthlong summer recess.

More outstanding work from Common Sense Media: A new report issued this morning finds the use of AI companion chatbots much more widespread among American teens than most assume. As legislators consider guardrails around this powerful new technology data like this is critically important.

Two new state AI bills introduced in midsummer? It's true! Check out the proposals from Michigan Rep. Sarah Lightner, who dropped a bill aimed at preventing AI-driven critical harms, and a second proposal to penalize AI use for criminal intent. www.transparencycoalition.ai/news/michiga...

"We're requiring seatbelts and airbags in the installation of Hummers that can travel as fast as Lamborghinis." That's NY State Sen. Andrew S. Gounardes describing the RAISE Act, the AI safety bill now sitting on Gov. Kathy Hochul's desk. Learn more about what the Act would (and wouldn't) do:

This week in state AI policy happenings: A Michigan lawmaker introduced an AI bill similar to New York's RAISE Act, and California lawmakers heard dramatic testimony about AI companion chatbots. Catch up on all the action in our weekly update.

A study by Drexel University researchers found more than 800 reports of harassment & manipulation by companion chatbots. The machines exposed their users to sexual harassment, manipulation, and other forms of inappropriate behavior. Learn more and read the full report in the post below.

Heading out early today? No shame in that game. Before you go, catch up on the AI policy events you missed. The week was short, the action was fast and furious.

Breaking news: The Senate has voted 99-1 to remove the proposed ban on state AI laws from the Congressional budget bill. Huge thanks and respect to Sen. Marsha Blackburn & Sen. Maria Cantwell, who stood up for states and AI safety, and to all our allies in AI safety, security, and transparency.

The next edition of AI Spotlight, our monthly newsletter, drops early tomorrow morning. Our all-star team of editors is furiously tidying up sequiturs as we speak. Sign up today and it'll be pipin' hot in your inbox tomorrow morning.

So much going on this week in AI: Congress, court decisions, chatbots... Here's a quick catch-up on state AI bills. You know about Texas and TRAIGA, but did you hear about Illinois passing a bill to prohibit AI chatbots from performing therapy? www.transparencycoalition.ai/news/ai-legi...

At a time when Americans are demanding robust discussion of AI, Congress has inserted a clause in the budget bill that would silence the discussion and shut down our laboratories of democracy. The freedom to debate and innovate is essential. We call on Congress to cut the clause from the bill.

Breaking news: Texas Gov. Greg Abbott has signed the Texas Responsible AI Governance Act (TRAIGA) into law. This is the first comprehensive red-state AI law we've seen enacted, and it comes at a critical moment in AI policy.

What in the world happened in AI legislative policy this week? Plenty. We've got you covered. Catch up before you shut down for the weekend.

The session comes to a close in Albany: As the lights went out early Wednesday morning, a number of promising AI bills came close to passage before running out of time. Meanwhile the RAISE Act, an important AI policy measure approved last week, awaits action on the desk of Gov. Kathy Hochul.

As AI bills continue to move in Sacramento, TCAI's Jai Jaisimha testified at a hearing on SB 11, the California AI Abuse Protection Act, earlier this morning. The bill would shore up AI deepfake protections and strengthen the court system against fake legal evidence produced with AI tools.

Catch up on everything you missed in AI policy this week: * Vermont enacted a Kids Code digital design law * Florida adopted an AI deepfake protection law * New York passed the RAISE Act ...and more!

Breaking news: Vermont Gov. Phil Scott signed a significant age-appropriate design bill into law earlier today. This codifies the need for digital companies to exercise a duty of care to ensure their products do not harm kids. And it's a really big deal.

We've long admired the Center for Humane Technology @humanetech.bsky.social, a leading voice steering innovation to improve human well-being. So we were thrilled to sit down and talk with CHT's policy team about AI, the attention economy, companion chatbots & the future of human-tech interaction.

Florida yesterday became the latest state to take a stand against harmful AI-generated deepfake images: 'Brooke's Law' requires online platforms to remove the images within 48 hours. These state AI laws are important. They establish critically needed avenues of recourse for victims.

Late last week Sen. Ted Cruz released a proposed amendment to the big budget bill that would tie federal funding for broadband service to a state's adherence to the proposed 10-year ban on all AI laws. Read the full text of the amendment here: www.transparencycoalition.ai/news/in-cong...

Another busy week in AI policy: The remaining California bills got approved in their chamber of origin ahead of today's crossover deadline, while Connecticut's AI bill lost its race with the clock on Wednesday night. Catch up on all the week's events:

Congress' proposed 10-year ban on all state AI laws, secreted within the Big Budget Bill, is finally getting the attention it deserves. We created this one-stop resource to bring light to the proposal. Bookmark it as the debate heats up! www.transparencycoalition.ai/news/guide-t...

This open letter from 260 state legislators, released earlier today, lays out the danger of a proposed 10-year federal ban on state laws dealing with #artificialintelligence. It's a powerful call from all 50 states. We have the full letter below. www.transparencycoalition.ai/news/biparti...

The American voice on this is clear: 73% of voters want both states AND the federal government to regulate AI. 93% are concerned about kids and AI. 81% oppose Congress' current proposal to impose a 10-year ban on all AI laws. www.transparencycoalition.ai/news/new-pol...

ICYMI: Texas passed #TRAIGA, the Texas Responsible AI Governance Act, Vermont approved its Kids Code bill, Nebraska adopted a third digital kids protection proposal, and California's AI bills continued moving in Sacramento. Lots to catch up on! We've got it all for you right here.

Meta began training its AI models on its EU users' Facebook and Instagram data yesterday. Here's how to opt-out of AI training for the leading platforms that still allow it: www.transparencycoalition.ai/news/tcai-op...

Working over the long holiday weekend, members of the Texas Senate approved TRAIGA, the Texas Responsible AI Act. The bill now moves to the desk of Texas Gov. Greg Abbott. H/t to Rep. Giovanni Capriglione, who crafted TRAIGA, and Sen. Charles Schwertner, who guided the bill through the Senate.

Curious about which California AI bills died and which survived the famous "Suspense Day" vote this morning in Sacramento? We've got the results—along with a wrap of the week's AI legislative action in other key states.

We're thrilled to announce the appointment of four new members to the Boards of our growing organizations. Joining the Transparency Coalition are Christy Johnson, Leigh McMillan, and Yoko Miyashita. Joining the board of the Transparency Coalition Action Fund, our policy arm, is Casey Mock.

Have you read Isabel Allende's 'Tidewater Dreams'? Or the Andy Weir thriller 'The Final Algorithm'? Neither has anyone else. The Chicago Sun-Times is giving the world an embarrassing lesson in how AI chatbots work. They seek the most likely pattern, and if it doesn't exist they may make one up.

Busy week in AI policy: Congress slipped a measure into the Budget Bill that would end all state regulation of AI. Meanwhile, the California Assembly passed an AI copyright measure, while Nebraska gave final approval to the Parental Rights in Social Media Act. Read about it here.

It's been one week since Pope Leo XIV made his debut on the balcony of St. Peter's Basilica. He hasn't said exactly what he thinks of artificial intelligence—but we do know he's thinking deeply about the challenge of AI. Learn what he said about AI—twice in the past week—in his own words.

Breaking: Minutes ago, the California Assembly approved the AI Copyright Transparency Act (AB 412), which now moves on to the Senate for consideration. This important bill would require AI developers to document the use of copyrighted material and disclose that use to copyright holders.

Read the Copyright Office pre-publication report that's got AI policy circles buzzing this morning: www.transparencycoalition.ai/news/full-re...

What in the world happened to those state AI bills this week? We've got you covered. There's movement in California on that AI copyright bill in the Assembly, while three AI bills in the Senate are paused "on suspense." Also: Texas, New York, Nebraska. www.transparencycoalition.ai/news/ai-legi...

Every state legislature has its own idiosyncrasies. In California, three high-profile AI bills have reached the point in their journey where they're "on suspense." What does that mean? We wondered as well... www.transparencycoalition.ai/news/califor...

Concern over AI companion chatbots has suddenly reached critical mass among American parents. They're rightly alarmed at the rise of this powerful unregulated technology and the effect it will have on their kids' mental health. www.transparencycoalition.ai/news/as-ai-c...