This week has certainly been amazing in many ways. First of all, we have had a Microsoft and Google event to present their respective AI integrated into search engines. But the best of all is that despite the fact that we are used to the fact that the prize always goes to Google, this time it has not been like that. But it has not only been our impression, but also the mistakes they have made have led them to lose 100 million dollars in the stock market.
Bard’s presentation event that took place this Tuesday can be summed up with the word ‘unnecessary’. A date in which they limited themselves to count less than what they already counted in a press release a few days ago without anything new, and on top of that with errors in its chatbot in which it was seen that it shared inaccurate information. These two factors made Google look like a company that had to pull something off to compete against Microsoft and its integration with ChatGPT.
‘Sgroogled.com’: when MICROSOFT launched ANTI-GOOGLE ads
Google suffers the error of its new AI in the pocket
This error was first noticed by Reuters, which published its article at 2:42 p.m. according to its web cache. And just minutes later you can see a collapse in the value of the shares in Google by almost 10 dollars per share. This fully affected the value of the company that was reduced by an inaccuracy of its AI by 100 billion dollars, and all thanks to the influence of a medium. In addition, this time also coincided with the end of his presentation event, which was disappointing by not having anything new and limiting himself to reading (with less information) his press release from several days ago.
The error that exposed Google is related to the James Webb Space Telescope. And it is that in a promotional video you can see how the question is formulated: What new discoveries from the James Webb Space Telescope (JWST) can I tell my 9-year-old son?
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 → https://t.co/fSp531xKy3 pic.twitter.com/JecHXVmt8l
—Google (@Google) February 6, 2023
Among the responses that are generated is that the telescope was used to take the first images of an exoplanet (planet outside the Solar System). But this is not correct, since the first images of exoplanets were taken in 2004 by the Very Large Telescope (VLT) from the European Southern Observatory according to NASA.
Although there are certain nuances to this error, since Bard is not wrong on a technical level. As The Financial Times points out, the telescope took the first image of the planet LHS 475b outside our Solar System, so it was the first exoplanet captured by the JWST and the first photograph of this planet. That is why the error itself can have nuances.
The NASA/ESA/CSA James #Webb Space Telescope has confirmed the presence of an exoplanet for the first time. Formally classified as LHS 475 b, the planet is almost exactly the same size as our own, clocking in at 99% of Earth’s diameter.
👉 https://t.co/3KPq5ynh5H#JWST pic.twitter.com/xU7grtICCl
— ESA (@esa) January 11, 2023
Although, in the face of this confusion, Google has not withdrawn the tweet nor has it qualified the response that Bard has offered at the time of writing this article. This makes one think that AIs cannot be trusted blindly, and that these can also be wrong as we are seeing from the hand of Google. What is clear is that being in a hurry is not a good companion, and right now we are facing a race between the big technology companies to find out who occupies the first place in AI implementation.