USA TODAY International Edition
Scientists: Keep grip on new tool
Open letter seeks mitigation of dire risks, including extinction
Hundreds of scientists, tech industry execs and public figures – including leaders of Google, Microsoft and ChatGPT – are sounding the alarm about artificial intelligence, writing in a new public statement that fast- evolving AI technology could create as high a risk of killing off humankind as nuclear war and COVID- 19- like pandemics.
“Mitigating the risk of extinction from AI should be a global priority alongside other societal- scale risks such as pandemics and nuclear war,” said the one- sentence statement, which was released by the Center for AI Safety, or CAIS, a San Francisco- based nonprofit organization.
CAIS said it released the statement as a way of encouraging AI experts, journalists, policymakers and the public to talk more about urgent risks relating to artificial intelligence.
Among the 350 signatories of the public statement were executives from the top four AI firms, OpenAI, Google DeepMind, Microsoft and Anthropic. One of them is renowned researcher and “Godfather of AI” Geoffrey Hinton, who quit his job as a vice president of Google so he could speak freely of the dangers of a technology he helped develop.
Also signing the statement: Sam Altman, the chief executive of OpenAI, the firm behind the popular conversation bot ChatGPT, which has made AI accessible to millions of users and allowed them to pose questions to it. Demis Hassabis, who heads Google’s AI division, also signed the statement.
Altman, Hinton and other industry leaders have become increasingly vocal about their concerns about AI and the need for some kind of technological guardrails for it, including government regulation.
“I think if this technology goes wrong, it can go quite wrong,” Altman said in a recent Senate Judiciary subcommittee hearing about potential oversight of AI. “And we want to be vocal about that. We want to work with the government to prevent that from happening.”
The CAIS statement allows readers to view signatories based on whether they are AI Scientists or “other notable figures.” The non- scientists include noted legal scholar Laurence Tribe of Harvard University, longtime climate change activist Bill McKibben and musician and artist known as Grimes, who shares two children with Tesla and Twitter leader Elon Musk. Musk himself is not one of the signatories.
Why is AI so alarming?
When he quit Google, the 75- year- old Hinton said he had grown increasingly uncomfortable with the rapid advances in AI technology, including its ability to show early indications of being able to develop simple cognitive reasoning.
Asked by a conference panel moderator what was the “worst case scenario that you think is conceivable,” Hinton replied without hesitation. “I think it’s quite conceivable,” he said, “that humanity is just a passing phase in the evolution of intelligence.”
In his Senate testimony two weeks ago, Altman told senators that one potential concern would be if AI can develop the ability to “self replicate and self-extricate into the wild.”
In Altman’s submitted written testimony for that hearing, he disclosed that external AI safety experts invited to review its latest chatbot version “helped identify potential concerns with GPT- 4 in areas including the generation of inaccurate information ( known as ‘ hallucinations’), hateful content, disinformation, and information related to the proliferation of conventional and unconventional weapons.”
Do all experts agree?
No. Some cybersecurity experts described Hinton’s comments after quitting Google as being extreme and unwarranted, including Michael Hamilton, a co- founder of the Critical Insight risk management firm and former vice chair of the Department of Homeland Security’s State, Local, Tribal and Territorial Government Coordinating Council.
Hamilton downplayed Hinton’s concerns in an interview with USA TODAY, saying that artificial intelligence is basically an extremely sophisticated programming platform that can only go so far as humans allow it. It cannot, for instance, become the kind of sentient, self- aware and all- knowing technology that created the Terminator of movie fame, Hamilton said.
And Altman, Hinton and other AI leaders have praised the huge potential AI has for helping humanity in a broad array of areas, from industry and education to improving the ability of disabled people to see, hear and communicate.
How did open letter come about?
The letter was posted by CAIS, which states that it believes that artificial intelligence “has the potential to profoundly benefit the world, provided that we can develop and use it safely.”
“Current systems already can pass the bar exam, write code, fold proteins, and even explain humor,” CAIS says on its website. “Like any other powerful technology, AI also carries inherent risks, including some which are potentially catastrophic.”
CAIS said when posting the new statement that many basic problems in AI safety have yet to be solved, and its mission is to reduce societal- scale risks associated with artificial intelligence by conducting safety research, building the field of AI safety researchers, and advocating for safety standards.
What is being done about AI?
Congress has held numerous hearings on the issue, calling in the top executives of the big AI firms.
At the recent Senate Judiciary subcommittee hearing, the chairman – Sen. Richard Blumenthal, D- Conn. – said he asked an AI chatbot to go through his floor speeches and deliver his opening statement for him. In sound and substance, Blumenthal noted, the chatbot mimicked his voice, and his concerns, precisely.
On May 4, following a White House meeting with tech leaders, the Biden administration announced a series of steps, including $ 140 million in new research efforts, to promote responsible innovation in the field of artificial intelligence and protect people’s rights and safety.
Under the new Biden initiative, the National Science Foundation will launch seven new National AI Research Institutes to bring together federal agencies, private- sector developers and academia in pursuing ethical and responsible development of AI that serves the public good. The new Institutes will advance AI research and development in critical areas, including climate change, agriculture, energy, public health, education and cybersecurity.