Joystiq
Joystiq News
If the unofficial debut of is any indication, . The company posted an ad to Twitter this week showing off the natural-language AI model displaying false information about the James Webb Space Telescope (JWST).
In the (via ), a short GIF shows an example of a Q&A with Bard. “What new discoveries from the James Webb Space Telescope can I tell my 9-year old about?” the query reads. The machine quickly spits out three ideas, including the last one that says, “JWST took the very first pictures of a planet outside of our own solar system. These distant worlds are called ‘exoplanets.’ Exo means ‘from outside.’” Although the bit about exoplanets is spot-on, the first part saying the JWST took the first pictures of them is false. That honor belongs to the European Southern Observatory’s Very Large Telescope (VLT) in 2004, as .
Although incorrect information in a Twitter ad won’t likely hurt anything directly, it’s easy to view the mistake as an omen of the risks of releasing natural-language chatbots into the wild. It parallels CNET’s decision to with an AI chatbot; they were also .
Because chatbots get so much right — and spit out answers with such supreme confidence — anyone who doesn’t fact-check their responses may be left with false beliefs. Considering the chaos that (non-AI-powered) misinformation has already let loose on society, releasing this often mind-blowing technology before it can be trusted to produce factual information reliably and accurately — even escaping Google’s copy editors — may have us in for a wild ride.
Console Bang News!
In the (via ), a short GIF shows an example of a Q&A with Bard. “What new discoveries from the James Webb Space Telescope can I tell my 9-year old about?” the query reads. The machine quickly spits out three ideas, including the last one that says, “JWST took the very first pictures of a planet outside of our own solar system. These distant worlds are called ‘exoplanets.’ Exo means ‘from outside.’” Although the bit about exoplanets is spot-on, the first part saying the JWST took the first pictures of them is false. That honor belongs to the European Southern Observatory’s Very Large Telescope (VLT) in 2004, as .
Bard is an experimental conversational AI service, powered by LaMDA. Built using our large language models and drawing on information from the web, it’s a launchpad for curiosity and can help simplify complex topics →
— Google (@Google)
Although incorrect information in a Twitter ad won’t likely hurt anything directly, it’s easy to view the mistake as an omen of the risks of releasing natural-language chatbots into the wild. It parallels CNET’s decision to with an AI chatbot; they were also .
Because chatbots get so much right — and spit out answers with such supreme confidence — anyone who doesn’t fact-check their responses may be left with false beliefs. Considering the chaos that (non-AI-powered) misinformation has already let loose on society, releasing this often mind-blowing technology before it can be trusted to produce factual information reliably and accurately — even escaping Google’s copy editors — may have us in for a wild ride.
Console Bang News!