California lawmakers want Governor Gavin News to approve the bills that they have adopted, which are aimed at making the chat -bots of artificial intelligence safer. But when the governor weighs whether the law should be signed in the law, he is faced with a familiar obstacle: objections from technological companies that say that new restrictions will impede innovation.
California companies are world leaders in AI and spent hundreds of billions of dollars to stay in front of the race to create the most powerful chat bots. A fast pace alarmed parents and legislators, concern that chat bots cause damage to the mental health of children, subjecting them to the content of self-confusion and other risks.
Parents who claim that the bots encourage their adolescents to harm themselves before they died of suicide, sued technological companies, such as Openai, personal technologies and Google. They also prompted a larger number of fences.
Calls for more regulation of AI were reflected throughout the capital of the country and various states. Even when the Trump administration “Action plan” It offers to reduce the red ribbon in order to encourage the development of AI, legislators and regulators of both parties solve the problems of children's safety related to chat bots that answer questions or act as digital companions.
California lawmakers this month adopted two accounts for the safety of AI chat bots with which the technological industry lobbied. Newsom must approve or reject them until mid -October.
The decision with high rates puts the governor in a difficult place. Politicians and technological companies want to assure the public that they protect young people. At the same time, technological companies are trying to expand the use of chat bots in classes and oppose new restrictions, which, according to them, go too far.
Suicide prevention resources and crisis consultations
If you or one of your friends fights suicidal thoughts, seek help from a professional and call 9-8-8. The first nationwide hot line in the field of mental health 988 in the United States 988 will connect subscribers with trained mental consultants. Text “House” 741741 in the USA and Canada to get to Crisis text lineField
Meanwhile, if Newsom is running for the presidency in 2028, he may need more financial support from wealthy technical entrepreneurs. On September 22, Newsom promoted the state’s partnerships with technological companies regarding AI efforts and advertised how the technological industry fed the economy of California, calling the Epicenter of American Innovations.
In the past, he put a veto on the legislation on the security of AI, including check Last year, this divided the technological industry of the Silicon Valley, because the governor thought that this gave the public a “false sense of security”. But he also made it clear that he was trying to find a balance between solving security problems and ensuring that California technological companies continue to dominate AI.
“We have a sense of responsibility and accountability to lead, so we support risk, but not recklessness,” News told News in discussion With former President Clinton at the event on the global initiative of Clinton on Wednesday.
Two bills sent to the governor – a bill on the Assembly of 1064 and the bill of Senate 243 – strive to make AI chat bots safer, but are faced with a rigid opposition from the technological industry. It is unclear whether the governor will sign both bills. His office refused to comment.
AB 1064 forbids a person, business and another organization from the creation of companions-bots available to a resident of California under the age of 18 years, unless a chat both is “foreseen” to harmful behavior, such as encouraging a child to self-confusion, violence or random nutrition.
SB 243 requires that the operators of the chat bots notify certain users that virtual assistants are not people.
In accordance with the bill, chatbot operators would have to have procedures to prevent suicide production or the maintenance of self-confinement and supply of fences, such as the direction of users to the hotline of suicide or the text line of the crisis.
They would have to notify insignificant users, at least every three hours to take a break, and that the chat is not a person. Operators will also be required to implement “reasonable measures” to prevent the creation of obvious content in sexual terms.
Tech Lobby Group Technet, whose members include Openai, Meta, Google and others, said in her statement that “agrees with the intention of bills,” but remains against them.
AB 1064 “imposes vague and non -working restrictions that create wide legal risks, while cutting out students from valuable AI training tools,” said Robert Boykin, Technet Executive Director in California and South -Wall. “SB 243 establishes more clear rules without blocking access, but we still have problems with its approach.”
The representative of META said that the company has “unintentional consequences that will have indicators such as AB 1064.” The technical company has launched a new Super Pak To combat state regulation, which the company considers too burdensome, and strives for greater parental control over how children use artificial intelligence, Axios He reported on Tuesday.
Opponents led by the computer and communication industry ASSN. AB 1064 aggressively lobbying, saying that this threatens innovation and does not affect California companies that would face a large number of trials and should decide whether they want to continue to work in the state.
Groups of advice, including common sense of the media, a non -profit organization that sponsored AB 1064 and recommends that minors not use AI companions, urge NEWSOM to sign a bill. California Atti. General Rob Bont also supports the score.
The Electronic Frontier Foundation said that SB 243 is too wide and is faced with free speech problems.
Several groups, including common sense in the media and technological supervision in California, deleted their support SB 243 after amendments were made to the bill, which, according to them, weakened the defense. Some of the changes in Limited, which receive certain notifications and included exceptions for certain chat bots in video games and virtual assistants used in intellectual speakers.
Legislators, which introduced the security legislation of the chat bots, want the governor to sign both bills, claiming that they can “work in harmony.”
Senator Steve Padilla, who introduced SB 243, said that even with the changes that he still believes that the new rules will become safer.
“We have a technology that has a great potential for good, incredibly powerful, but develops incredibly fast, and we cannot miss the window to ensure that they are fences of common sense to protect people,” he said. “I am satisfied with where the account is.”
Assembly Rebecca Bauer-Kahan, who wrote AB 1064, said that its bill balances the advantages of artificial intelligence, protecting against dangers.
“We want to make sure that when children interact with which chat -bout, this does not create unhealthy emotional affection, directing them to suicide, random nutrition, any things that, as we know, are harmful to children,” she said.
During the legislative session, legislators heard from mourning parents who lost their children. AB 1064 emphasizes two high -profile trials: one against the manufacturer San Francisco Chatgpt Openai and the other against the characters of Technologies, the developer Platform Platform.ai.
Character This is a platform where people can create and interact with digital characters that imitate real and fictional people. Last year, in the federal lawsuit of Florida, Mom Garcia claimed that the chat -bouts of the character. The mental health of her son Sewell network III was damaged and accused the company of not notifying her or offered help when he expressed thoughts of suicide virtual characters.
More families sued the company this year. A Character The press secretary said that they very deeply care about the safety of users and “encourage legislators to properly create laws that contribute to the safety of users, and also allow enough space for innovation and free expression.”
In August, California parents of Adam Rhine sued Openai, claiming that Chatgpt provided the teenager with information about suicide methods, including the one that the teenager used for himself.
Openai said that she strengthens guarantees and plans to release parental control. Its Executive Director Sam Altman wrote in the September blog that the company believes that minors need “significant protection”, and the company distributes “safety before confidentiality and freedom for adolescents”. The company refused to comment on California accounts for the chat bot.
For California legislators, the clock is ticking.
“We are doing our best,” Bauer-Kahan said. “The fact that we have already seen how children lose their lives from the AI tells me that we are not moving fast enough.”
