The world of AI-powered technology is trending and that means we’re seeing chatbots sprout up in every nook and corner.
From Google to Microsoft, all tech giants are offering their rivals against OpenAI’s ChatGPT. But the real question is how reliable are these tools, to begin with.
Well, Google’s chatbot called Bard is preparing for internal testing and a recent demo showcased some rather shocking findings. It displayed plenty of inaccuracies and that means you might not want to put blind trust in any chatbot after all.
Bard is Google’s new offering which the firm went public this week. It’s all set to arrive in the next few weeks and to help market the offering better, the search engine giant was seen adding a GIF across social media. This showed Bard putting solutions to queries about new offerings made by a leading NASA telescope.
Moreover, the program enlisted the answers and made it awfully simple for anyone to read and understand through bullet points. This is also where the company can easily pick up facts and won’t be scrolling through piles and piles of search results that end up finding a suitable website.
So if everything was going according to plan, where did the errors arise? Well, plenty of specialists in astronomy claim there were some huge inaccuracies in the findings that were laid out by Bard. This program wrongly says the telescope was the first to click images of a planet located out of the solar system. But in all reality, it was the astronomer who clicked the first picture.
Meanwhile, Bard seems to have made a huge error but when you come to think of it, the telescope did take an image of the planet but it was not the first one to do so. So NASA’s press statements on the discovery may have made the program confused.
Whatever the case may be, one thing is for sure. It’s definitely embarrassing to see something like this because the whole use of AI-powered chatbots is to enhance a user’s overall search experience. But with something like this arising, you can’t help but wonder that it just might not be ready for release to the masses.
And the Android maker knows this. It mentioned through one of its live streams today from Paris that it has no intention of launching the chatbot to the public yet because it’s undergoing internal testing. This means it’s far from ready. Only when the company feels like the quality of results meets a particular standard or criteria that is satisfactory, would a release be mentioned.
The approach is awfully different from that taken on by fellow competitor Microsoft and its respective AI-powered Bing chatbot.
All in all, this demo has opened many people’s eyes to what AI-powered technology holds for the future and how programs setting out information through such means might need to be double-checked for authenticity more than anything else.
Read next: Google Unveils Exciting New Features For Its Maps, Search And Translate Services
From Google to Microsoft, all tech giants are offering their rivals against OpenAI’s ChatGPT. But the real question is how reliable are these tools, to begin with.
Well, Google’s chatbot called Bard is preparing for internal testing and a recent demo showcased some rather shocking findings. It displayed plenty of inaccuracies and that means you might not want to put blind trust in any chatbot after all.
Bard is Google’s new offering which the firm went public this week. It’s all set to arrive in the next few weeks and to help market the offering better, the search engine giant was seen adding a GIF across social media. This showed Bard putting solutions to queries about new offerings made by a leading NASA telescope.
Moreover, the program enlisted the answers and made it awfully simple for anyone to read and understand through bullet points. This is also where the company can easily pick up facts and won’t be scrolling through piles and piles of search results that end up finding a suitable website.
So if everything was going according to plan, where did the errors arise? Well, plenty of specialists in astronomy claim there were some huge inaccuracies in the findings that were laid out by Bard. This program wrongly says the telescope was the first to click images of a planet located out of the solar system. But in all reality, it was the astronomer who clicked the first picture.
Meanwhile, Bard seems to have made a huge error but when you come to think of it, the telescope did take an image of the planet but it was not the first one to do so. So NASA’s press statements on the discovery may have made the program confused.
Whatever the case may be, one thing is for sure. It’s definitely embarrassing to see something like this because the whole use of AI-powered chatbots is to enhance a user’s overall search experience. But with something like this arising, you can’t help but wonder that it just might not be ready for release to the masses.
Speaking as someone who imaged an exoplanet 14 years before JWST was launched, it feels like you should find a better example?
— Bruce Macintosh (@bmac_astro) February 8, 2023
And the Android maker knows this. It mentioned through one of its live streams today from Paris that it has no intention of launching the chatbot to the public yet because it’s undergoing internal testing. This means it’s far from ready. Only when the company feels like the quality of results meets a particular standard or criteria that is satisfactory, would a release be mentioned.
The approach is awfully different from that taken on by fellow competitor Microsoft and its respective AI-powered Bing chatbot.
All in all, this demo has opened many people’s eyes to what AI-powered technology holds for the future and how programs setting out information through such means might need to be double-checked for authenticity more than anything else.
Read next: Google Unveils Exciting New Features For Its Maps, Search And Translate Services
