Bard misled Google's announcement.
Big Tech is rushing to produce products to compete with OpenAI's ChatGPT, the fastest-growing app ever. According to a Google blog post, Bard will be published "in the coming weeks(Opens in a new tab)" but is already imitating ChatGPT by creating false information.
In Google's Bard blog post, an animated graphic shows the Bard user experience, however the AI incorrectly says the James Webb Space Telescope took the first exoplanet photo. Google tweeted this animation.
Webb took its first exoplanet photo last September (Opens in a new tab), but 2004 was the first (Opens in a new tab).
Bard's assertion concerning James Webb is recent, which raises questions. Language models use complex sentence-completion techniques to generate information. An AI may make more mistakes in sentences concerning the recent past because the material hasn't been written as often. This may be why ChatGPT's model won't tell you much after 2021. (Opens in a new tab). https://ejtandemonium.com/
This mishap highlights the issue of all generative AI companies ignoring truth value, which may encourage consumers to use old-fashioned search engines. Microsoft is adding a ChatGPT-like answer engine to Bing, but "Bing will sometimes distort the information it finds, and you may receive responses that sound plausible but are incomplete, wrong, or unsuitable," according to Microsoft(Opens in a new tab). http://sentrateknikaprima.com/
Big Tech is rushing to produce products to compete with OpenAI's ChatGPT, the fastest-growing app ever. According to a Google blog post, Bard will be published "in the coming weeks(Opens in a new tab)" but is already imitating ChatGPT by creating false information.
In Google's Bard blog post, an animated graphic shows the Bard user experience, however the AI incorrectly says the James Webb Space Telescope took the first exoplanet photo. Google tweeted this animation.
Webb took its first exoplanet photo last September (Opens in a new tab), but 2004 was the first (Opens in a new tab).
Bard's assertion concerning James Webb is recent, which raises questions. Language models use complex sentence-completion techniques to generate information. An AI may make more mistakes in sentences concerning the recent past because the material hasn't been written as often. This may be why ChatGPT's model won't tell you much after 2021. (Opens in a new tab). https://ejtandemonium.com/
This mishap highlights the issue of all generative AI companies ignoring truth value, which may encourage consumers to use old-fashioned search engines. Microsoft is adding a ChatGPT-like answer engine to Bing, but "Bing will sometimes distort the information it finds, and you may receive responses that sound plausible but are incomplete, wrong, or unsuitable," according to Microsoft(Opens in a new tab). http://sentrateknikaprima.com/
