"The best lesson for artificial intelligence may be Thursday's "rapid unscheduled disassembly" of SpaceX's Starship rocket, aborted four minutes after launch. With ChatGPT prompting speculation about mankind's destruction, you should know that techies have obsessed seemingly forever over what's known as the Paper Clip Theory -- the idea that if you told an artificial-intelligence system to maximize the production of paper clips, it would fill the whole world with paper clips. Another version of the theory, Strawberry Fields Forever, has AI using every piece of available dirt to grow strawberries. Scary, right? So are "Halloween" movies.
Not to be outdone, decision theorist (huh?) Eliezer Yudkowsky recently wrote in Time magazine that the "most likely result of building a superhumanly smart AI" is that "literally everyone on Earth will die." Literally everyone! That's ludicrous, as is most clickbait these days. Sam Altman, CEO of ChatGPT creator OpenAI, told podcaster Lex Fridman, "There is some chance of that." C'mon now.
Apparently, Pandora's box has opened and is spewing its evils, which ignores all the good uses of large language models that will transform software coding, the workplace, education and more. Sadly, our geniuses in government appear to be the remaining few who still read Time magazine. So bring on the regulators to shut the box heroically.
Earlier this month, the Commerce Department initiated the process of regulating artificial intelligence, with Assistant Secretary Alan Davidson suggesting, "We know that we need to put some guardrails in place to make sure that they are being used responsibly." Bad idea. Guardrails are for children bowling at birthday parties. AI is in its infancy, and we don't yet know how it will change industries and society. Don't freeze it now.
If the U.S. regulates AI, research will just move somewhere that doesn't regulate it, maybe the Bahamas, where the unkempt coders of the future could keep cranking away. Or worse, China.
Google CEO Sundar Pichai told CBS's "60 Minutes" that he wants global AI regulation.
Elon Musk and a group of AI experts wrote an open letter calling for an immediate six-month pause of "giant AI experiments." Isn't it interesting that those who need to catch up are pushing for a pause?
We don't need onerous new regulations. There are already laws on the books for false advertising, copyright infringement and human endangerment. Do we really need bureaucrats who still use machines programmed in outdated Cobol to create regulations for a nascent technology they don't understand?
But to assuage worries, I would recommend one tiny rule for AI development: Include a kill switch.
Every rocket launch has a kill switch -- an abort button that explodes the rocket before it endangers humans. The only sensible regulation for AI is to demand a similar kill switch. For consumer use, maybe it's a safe word like "Bosco" or a phrase like "Open the pod bay doors, HAL" that would immediately stop the AI actions -- a virtual way to unplug the machine. Data centers where much of the AI computation is done, such as Amazon Web Services and Microsoft Azure, could easily implement such switches.
We've always had the ability to stop computers: Hitting Ctrl-C halted programs on early computers. Windows PCs have Ctrl-Alt-Delete. I often scream "Stop!" at Alexa. Let's keep humans in control. Mr. Altman suggests OpenAI has implemented a kill switch. Smart.
Most AI systems are built with safety and alignment -- meaning aligned with human welfare. Mr. Altman stresses their importance. Can you regulate that? I doubt it. But regulations could (or likely will) become political tools, like the Biden administration's short-lived Disinformation Governance Board. China is instituting rules, stating that AI should "reflect the core values of socialism, and must not contain subversion of state power." I don't think the free world wants to emulate that.
Can we trust the technology industry and its ethicists to get it right? We saw gene-editing ethics violations denounced immediately in 2018 when Chinese scientist He Jiankui claimed to have altered the genomes of twin girls to make them resistant to HIV. Expect the same reactions with AI.
Those wanting to halt progress will wail about the existential threat to humans from AI. Does that sound familiar? We had nuclear winter, the Bermuda Triangle, global cooling then warming then change, Donald Trump and now AI death bots.
There are always people who need threats hanging over their lives to get out of bed in the morning. Fine, let them worry. But innovation-dampening regulations aren't the answer. Rockets fail, so will AI systems. I'd feel safer with a kill switch than with regulations." [1]
1. Inside View: AI Simply Needs a Kill Switch
Kessler, Andy. Wall Street Journal, Eastern edition; New York, N.Y. [New York, N.Y]. 24 Apr 2023: A.15.
Komentarų nėra:
Rašyti komentarą