<p>Last week, in a surprise turn of events, Google and Microsoft on short notice to the media, hosted back-to-back events to showcase their new AI (Artificial Intelligence)-powered products.</p>.<p>Google won the race over Microsoft by a day (February 7) to reveal the Bard, the improved AI language model that would enhance its search engine. It is capable of delivering insightful answers be it a general knowledge question like knowing the height of the Himalayas or a complex query like suggesting a one-week trip itinerary to an exotic place.</p>.<p>Google even played a demo video showing Bard's capability of teaching a nine-year-old kid on NASA’s James Webb Space Telescope (JWST).</p>.<p>It listed several easy-to-understand pointers, but one was factually wrong. It claimed JWST was the first to take pictures of planets outside the solar system. But, they were actually captured by the European Southern Observatory’s Very Large Telescope (VLT) in 2004.</p>.<p>The media, though took a day to notice the blunder, did not spare Google. This led to a PR nightmare for Google which apparently lost more than $100 billion in market value.</p>.<p>Now, it has come to light that Microsoft ChatGPT-powered Bing also gave out factually wrong answers to queries.<br /><br />During the demo on February 8, Bing was asked to analyze the Q3 2022 financial reports of Gap and Lululemon and deliver short summaries. It inaccurately showed that Gap reported an adjusted gross margin of 37.4 per cent including the impairment charges. But, in actuality, it was 38.7 per cent.</p>.<p>Also, it even conjured up nonexistent data that Gap saw its operating margin grow by 5.9 per cent. </p>.<p>"Gap Inc. reported operating margin of 5.9%, adjusted for impairment charges and restructuring costs, and diluted earnings per share of $0.42, adjusted for impairment charges, restructuring costs, and tax impacts."- as shown in Bing's summary.</p>.<p>This data was not at all mentioned in the official Gap earnings document. </p>.<p>Even with regard to Lululemon, the earnings data churned up by <a href="https://www.deccanherald.com/business/technology/microsoft-bing-gets-openais-chatgpt-booster-shot-1189054.html" target="_blank">ChatGPT-powered Bing</a> were laced with inaccuracies.</p>.<p>The complex financial results are always tricky for commoners and regular journalists, and errors were not noticed until independent <a href="https://dkb.blog/p/bing-ai-cant-be-trusted" target="_blank">AI researcher Dmitri Brereton</a> went deeper to compare the search results of Bing and actual data.</p>.<p>Going by the embarrassing gaffes made by <a href="https://www.deccanherald.com/business/technology/google-announces-chatgpt-rival-ai-chatbot-bard-1188700.html" target="_blank">Google's Bard</a> and Microsoft-OpenAI's ChatGPT-powered Bing, the companies in their bid to outdo each other, seem to have rushed too fast to showcase their AI products.</p>.<p>It should not be which product came first, but the focus has to be on- delivering genuine information and offering a better user experience.</p>.<p>It would have been better if the companies trained the AI bot with more accurate information and conducted rigorous tests to ensure data delivered to the queries are genuine. Also, AI generative language models have been known to show human biases such as sexism, racism, and other prejudices too, and it is very important, they are ironed out and be objective and sensitive to controversial topics. </p>.<p>Currently, the Google search engine commands an envious 92+ per cent market share, and the rivals Microsoft Bing, Apple Safari, and others, together make up for less than seven per cent. </p>.<p>And, even with ChatGPT integrated with Bing and Edge browsers, the user interface and design layout of the search engine is not as visually engaging as Google search on Chrome.</p>.<p>It will be better for Google, Microsoft, and others to fully test their AI products before rolling out the feature to the public. Also, it will be great if they could come up with revenue sharing model with news and content-generating companies, so they can sustain the business.</p>.<p>If users get direct answers or news with short summaries on a dedicated AI chatbot tab, why would they ever visit a news website for information? Hope better sense prevails.</p>.<p><em>Get the latest news on new launches, gadget reviews, apps, cybersecurity, and more on personal technology only on <a href="https://www.deccanherald.com/tag/dh-tech?_ga=2.210580691.73733284.1595225125-1706599323.1592232366" target="_blank">DH Tech</a>.</em></p>
<p>Last week, in a surprise turn of events, Google and Microsoft on short notice to the media, hosted back-to-back events to showcase their new AI (Artificial Intelligence)-powered products.</p>.<p>Google won the race over Microsoft by a day (February 7) to reveal the Bard, the improved AI language model that would enhance its search engine. It is capable of delivering insightful answers be it a general knowledge question like knowing the height of the Himalayas or a complex query like suggesting a one-week trip itinerary to an exotic place.</p>.<p>Google even played a demo video showing Bard's capability of teaching a nine-year-old kid on NASA’s James Webb Space Telescope (JWST).</p>.<p>It listed several easy-to-understand pointers, but one was factually wrong. It claimed JWST was the first to take pictures of planets outside the solar system. But, they were actually captured by the European Southern Observatory’s Very Large Telescope (VLT) in 2004.</p>.<p>The media, though took a day to notice the blunder, did not spare Google. This led to a PR nightmare for Google which apparently lost more than $100 billion in market value.</p>.<p>Now, it has come to light that Microsoft ChatGPT-powered Bing also gave out factually wrong answers to queries.<br /><br />During the demo on February 8, Bing was asked to analyze the Q3 2022 financial reports of Gap and Lululemon and deliver short summaries. It inaccurately showed that Gap reported an adjusted gross margin of 37.4 per cent including the impairment charges. But, in actuality, it was 38.7 per cent.</p>.<p>Also, it even conjured up nonexistent data that Gap saw its operating margin grow by 5.9 per cent. </p>.<p>"Gap Inc. reported operating margin of 5.9%, adjusted for impairment charges and restructuring costs, and diluted earnings per share of $0.42, adjusted for impairment charges, restructuring costs, and tax impacts."- as shown in Bing's summary.</p>.<p>This data was not at all mentioned in the official Gap earnings document. </p>.<p>Even with regard to Lululemon, the earnings data churned up by <a href="https://www.deccanherald.com/business/technology/microsoft-bing-gets-openais-chatgpt-booster-shot-1189054.html" target="_blank">ChatGPT-powered Bing</a> were laced with inaccuracies.</p>.<p>The complex financial results are always tricky for commoners and regular journalists, and errors were not noticed until independent <a href="https://dkb.blog/p/bing-ai-cant-be-trusted" target="_blank">AI researcher Dmitri Brereton</a> went deeper to compare the search results of Bing and actual data.</p>.<p>Going by the embarrassing gaffes made by <a href="https://www.deccanherald.com/business/technology/google-announces-chatgpt-rival-ai-chatbot-bard-1188700.html" target="_blank">Google's Bard</a> and Microsoft-OpenAI's ChatGPT-powered Bing, the companies in their bid to outdo each other, seem to have rushed too fast to showcase their AI products.</p>.<p>It should not be which product came first, but the focus has to be on- delivering genuine information and offering a better user experience.</p>.<p>It would have been better if the companies trained the AI bot with more accurate information and conducted rigorous tests to ensure data delivered to the queries are genuine. Also, AI generative language models have been known to show human biases such as sexism, racism, and other prejudices too, and it is very important, they are ironed out and be objective and sensitive to controversial topics. </p>.<p>Currently, the Google search engine commands an envious 92+ per cent market share, and the rivals Microsoft Bing, Apple Safari, and others, together make up for less than seven per cent. </p>.<p>And, even with ChatGPT integrated with Bing and Edge browsers, the user interface and design layout of the search engine is not as visually engaging as Google search on Chrome.</p>.<p>It will be better for Google, Microsoft, and others to fully test their AI products before rolling out the feature to the public. Also, it will be great if they could come up with revenue sharing model with news and content-generating companies, so they can sustain the business.</p>.<p>If users get direct answers or news with short summaries on a dedicated AI chatbot tab, why would they ever visit a news website for information? Hope better sense prevails.</p>.<p><em>Get the latest news on new launches, gadget reviews, apps, cybersecurity, and more on personal technology only on <a href="https://www.deccanherald.com/tag/dh-tech?_ga=2.210580691.73733284.1595225125-1706599323.1592232366" target="_blank">DH Tech</a>.</em></p>