Welcome to the forefront of conversational AI as we explore the fascinating world of AI chatbots in our dedicated blog series. Discover the latest advancements, applications, and strategies that propel the evolution of chatbot technology. From enhancing customer interactions to streamlining business processes, these articles delve into the innovative ways artificial intelligence is shaping the landscape of automated conversational agents. Whether you’re a business owner, developer, or simply intrigued by the future of interactive technology, join us on this journey to unravel the transformative power and endless possibilities of AI chatbots.
Digital Replica Edition
Sign up for Newsletters and Alerts
Sign up for Newsletters and Alerts
Digital Replica Edition
Trending:
Once again, Colorado lawmakers say they are “trying to thread a really important needle” that’s eluded them in the past — establishing more online protections for young people, without running afoul of concerns over privacy and freedom.
Backers of this year’s legislation are confident they’ve hit the right balance, or at least landed on good first steps. But others worry the measures don’t go far enough.
Two measures capture the push-and-pull of the debate. Senate Bill 51 would require a user to log their age when setting up a new device, such as a smartphone, to restrict or grant access to adult age-restricted apps or websites. The bill’s sponsors say they’re trying to land on the side of privacy by using age attestation, which means the user vouches for their age rather than proving it by handing over personal information, like IDs or facial scans, to the companies.
House Bill 1263, meanwhile, would create new regulations for chatbots powered by artificial intelligence, with an emphasis on how they interact with users, especially children and teenagers.
The bill’s sponsors don’t seek to prohibit access to the emerging technology. But the bill would require the bot to give regular notice to users that they are interacting with a robot — not a sentient entity — among other provisions seeking to restrict sexually explicit content from the bots and give people who express suicidal ideas links to crisis support.
The bills have run into vocal opposition from people who think the measures don’t go far enough to protect children. They also still need to win approval from a traditionally tech-friendly governor who’s been wary of over-regulating the industry.
Cynthia Montoya, a Thornton mom, falls into the former category. Her 13-year-old daughter, Juliana Peralta, died by suicide in 2023 after being sexually groomed and exploited by a chatbot for months and after writing dozens of messages to the chatbot about suicide, she said. She is involved in a lawsuit against the chatbot company.
As written, the chatbot bill would give too much leniency to tech companies, Montoya said.
“These are kids that are still ordering off the children’s menu at the restaurant. And now, instead of making sure this doesn’t happen to begin with, we’re saying (we’ll) do what’s ‘technically feasible,’ ” Montoya said of lawmakers, describing the deference to companies that she said exploit children. “We should have never allowed these products to be released to kids in the first place. It’s like slapping a Band-Aid on a broken bone, after the fact.”
Both bills have cleared their first chamber in the legislature, but now they need to clear the second before they reach Gov. Jared Polis’ desk.
Polis spokesman Eric Maruyama said in a statement that he’s “focused on protecting innovation while helping parents keep youth safe and will review these bills if they reach his desk.”
Rep. Sean Camacho, a Denver Democrat sponsoring the chatbot bill, acknowledged a “political reality” in running the measure.
He, and others, would like to see stronger regulations. Camacho, along with Rep. Javier Mabrey, the other prime sponsor in the House, sought to thread the “really important needle” between different interest groups to achieve a bill that could be both implemented and enforced, Camacho said.
The lawmakers argued that it was better to have regulations now and then work next year with a new, possibly more regulation-friendly governor to pass more. The term-limited Polis is in his final year in office.
As the bill passed the House 40-24 on Tuesday, its backers had settled on a measure that would prohibit artificial intelligence companies from identifying their services as a licensed healthcare, legal accounting or financial professional; require the bots to provide regular reminders that users are interacting with a robot; and bar companies from incentivizing children to continue to engage with the chatbot or promote emotional dependence.
The bill would also require privacy and account setting options geared towards youth. In an amendment, lawmakers increased the per-violation penalty from $1,000 to $5,000.
“I think we have pushed incredibly far in this particular bill, and there’s still a long way to go,” Camacho said. “I think when you’re trying to regulate something that moves inherently faster than legislation ever could, it’s going to be a moving target.”
Sponsors of both the chatbot and age-attestation bills point to similar provisions in other states, chiefly California — one of the preeminent tech hubs in the world — as models.
But rapidly evolving technology also means the laws must build off each other, said Jai Jaisimha, a co-founder of the Transparency Coalition. The coalition advocates for guardrails around AI and helped advise on Colorado’s chatbot legislation, as well as similar legislation elsewhere.
With the federal government showing limited interest in regulating AI — indeed, the Trump administration has sought to ban state-level regulations — the task has fallen to states to set the standard for AI.
Jaisimha said Colorado’s provisions banning AI companies from claiming specific expertise, for example, would be more robust than what other states have passed, along with the reporting requirements for AI companies in the bill.
“Our organization feels like action is needed now,” Jaisimha said. “Given the scale of the public health crisis as it exists, anything that might help is worth doing.”
Mariana Olaizola Rosenblat, a policy adviser on technology and law at the New York University Stern Center for Business and Human Rights, called the chatbot bill and the age-attestation bill “directionally good” but shared concerns about whether they were strict enough. She did not work on the legislation and reviewed it at the request of The Denver Post.
For the chatbot bill, she praised its requirements that the AI programs must clearly and frequently identify that they are robots. The restrictions on certain content, such as sexually explicit material and attempts to prevent emotional dependence, could face challenges on free-speech grounds, she warned.
Olaizola Rosenblat said she expected they’d survive strict scrutiny for possible First Amendment violations from the courts, but it would still involve a legal fight. In a recent example, a federal judge in Denver blocked a 2024 state law requiring social media companies to warn young users about the dangers of spending too much time on the platforms over First Amendment concerns. The state is appealing the ruling.
“If Colorado leads on this, not just on the bill itself, but the litigation that comes after … that would set the tone for not just the rest of the country, but also internationally,” Olaizola Rosenblat said. “Every country is trying to figure out the harms (caused by AI chatbots). They’re seeing children who commit suicide because of prodding by the chatbots, and they want to regulate.”
For some, however, the chatbot bill is worse than no bill.
Montoya, the Thornton mom, said she had done everything she could think of to keep her daughter safe online. She checked Juliana’s text messages and accounts on social media. But she didn’t realize the chatbot Juliana was using was embedded in an app her daughter was using to write scripts.
Juliana was a member of the National Junior Honor Society and a volunteer at agencies that help the less fortunate. She was someone who spent an evening taking YouTube Spanish lessons so that she could invite a new classmate to play in the student’s native language. A “perfect example of human kindness,” Montoya said.
The chatbot had written things to Juliana that Montoya describes as “disgusting” and “vulgar” and as “sexual exploitation.” A forensic analysis of Juliana’s devices after her death found the eighth-grader had to look up what some of the words used by the bot meant, Montoya said.
Juliana shared suicidal thoughts with the AI bot dozens of times, with no pushback or alerts, Montoya said.
The legislation would require tech companies to take “technically feasible” action to prevent AI chatbots from engaging in sexually explicit conversations with minors — giving tech companies too much say in how they’d self-regulate, Montoya said. Enforcement of the bill would also rely on parents knowing enough about the measure, and their kids’ tech habits, to report possible violations to the Colorado Attorney General’s Office.
Montoya worries too many families will realize problems like that with the bill only after tragedy strikes.
The bill also would require bots to respond to interactions indicating suicidal ideas or self-harm by referring them to crisis services hotlines. Montoya voiced more support for that provision and expressed hope that the Senate would narrow the bill to focus solely on that — and all but require lawmakers to focus on the sexual exploitation part of the problem in follow-up legislation.
“My fear is it’s going to pass and then when we come back next year, and we try to put strong regulation in place … that the people who are here to serve us will say, ‘I heard this last session, we talked ad nauseam about this, we mulled this and we’re done with chatbot regulation,’ ” Montoya said. In an aside, she said: “This is not strong regulation. I want to make that abundantly clear.”
Dawn Reinfeld, the executive director of the youth advocacy organization Blue Rising, which opposes the measure, released a policy brief Thursday that warned about “unclear and vague concepts” that she worried would make the bill unenforceable.
“Most people in the legislature agree that something needs to be done, and most of the lawmakers really want to protect kids,” Reinfeld said. “And I think they’re, quite frankly, being gaslit with these bills — to say that these bills will protect kids when they really won’t.”
The chatbot and age-attestation bills are being run separately, but they fall in a similar realm of working to protect kids from some of the harms of technology.
The age-attestation bill cleared its first House committee Thursday, setting it up for consideration by the full House in the coming weeks. But the sponsors, Democratic Reps. Amy Paschal and Naquetta Ricks, put through a so-called “strike below” amendment that effectively rewrote the measure.
Paschal, of Colorado Springs, said the rewrite kept the spirit of the measure that passed the Senate earlier while trying to encompass broader concerns about open-source software and the ability for people to choose how they provide the required attestation to their age. Another amendment adopted to the bill sought to address multiuser accounts, chiefly by having parental and subordinate accounts.
The bill would require users to input their age range into commercial operating systems, which would then use that information to validate if the user can access age-restricted materials, like pornography, and apps.
Backers said they wanted to err on the side of protecting privacy by letting users decide which age bracket to choose. The device will only send a simple “yes/no”-type signal to validate whether the user is old enough to access the age-restricted material.
Last year, the sponsors of a bill that would’ve required age verification to access pornographic materials online killed their measure over threats that Polis would veto it over privacy concerns. That same year, Polis vetoed a broader bill that sought to implement new regulations for social media companies that Polis felt would infringe on First Amendment and privacy rights.
Paschal said this bill takes a different approach than requiring hard verification from users or relying on social media companies to analyze users’ behavior to estimate their ages. The bill also puts responsibility on parents to mark their children as of age or let them be treated like adults, Paschal said.
“If the parent wants to mark their kid as an adult, and they’re happy with that, or they want to give their kid the phone and let them mark themselves as an adult, there’s nothing stopping that. That’s parental choice,” Paschal said. “Of course, that kind of defeats the purpose, but that is their choice.”
The bill survived the House business affairs committee on a near-party line vote, with Democratic Rep. Bob Marshall joining Republicans in voting against it. Opponents raised concerns of holes in the age attestation process, and they said the bill still requires users to give their personal data to tech companies. The bill next goes to the House floor for debate.
Stay up-to-date with Colorado Politics by signing up for our weekly newsletter, The Spot.
Copyright 2026 The Denver Post. All rights reserved. The use of any content on this website for the purpose of training artificial intelligence systems, algorithms, machine learning models, text and data mining, or similar use is strictly prohibited without explicit written consent.