Alphabet shares dive after Google AI chatbot Bard flubs answer in ad
Alphabet Inc lost $100 billion in market value on Wednesday after its new chatbot shared inaccurate information in a promotional video and a company event failed to dazzle, feeding worries that the Google parent is losing ground to rival Microsoft Corp.
Alphabet’s stock took a dive during regular trading, declining as much as 9% and with trading volumes significantly higher than its 50-day moving average. However, the stock recovered slightly after hours, ending roughly unchanged. Despite losing 40% of its value in the previous year, the stock had seen a 15% increase since the start of the year, excluding Wednesday’s decline.
The error in Google’s ad for its newly launched chatbot Bard was first reported by Reuters. The advertisement incorrectly claimed which satellite took the first images of a planet outside the Earth’s solar system.
Google has been put on the defensive after OpenAI, a startup backed by Microsoft with around $10 billion, released software in November that has impressed consumers and captured the attention of Silicon Valley with its impressively accurate and well-written responses to simple questions.
Google’s presentation on Wednesday, which was live-streamed, did not provide information on when or how it plans to integrate Bard into its core search function. However, the day before, Microsoft held an event to announce that it has already made a version of its Bing search engine with integrated ChatGPT functions available to the public.
Read More: Google to enter the AI race with Microsoft
Google, based in Mountain View, California, discovered Bard’s error just before the presentation.
According to Gil Luria, a senior software analyst at D.A. Davidson, Google has failed to implement AI technology into their search product, despite leading the industry in AI innovation in recent years. Luria claims that Google has been playing catch-up in their search efforts and this haste led to a rushed announcement and the embarrassment of posting an incorrect answer during a demo.
The announcement caused Microsoft shares to rise by 3% on Wednesday, but remained unchanged in post-market trading.
Alphabet attempted to showcase their new AI feature, Bard, by posting a GIF video on Twitter, promising to make complex topics simple. However, the demonstration was marred by an inaccurate answer.
The advertisement showed Bard being asked to explain new discoveries from the James Webb Space Telescope to a 9-year-old. In response, Bard made a mistake by suggesting that the JWST took the first pictures of exoplanets, when in reality, the European Southern Observatory’s Very Large Telescope captured the first images in 2004, as confirmed by NASA.
Read More: China’s Alibaba joins global chatbot race
A Google spokesperson acknowledged the error and stated that they will be launching a Trusted Tester program this week to ensure Bard’s responses meet high standards for quality, safety, and accuracy. The program will involve both external feedback and internal testing.
FORMIDABLE COMPETITOR
Following a lackluster fourth quarter, Alphabet is striving to keep up with its competitors, including OpenAI. Reports indicate that the search and advertising giant has brought in co-founders Sergey Brin and Larry Page to expedite their efforts.
Some experts are questioning if Microsoft will become a formidable adversary to Google’s main business, however, King Lip, the chief strategist at Baker Avenue Wealth Management, suggests that these concerns may be overstated. Lip owns shares in both Alphabet and Microsoft and believes that Bing still has a long way to go before matching Google’s search capabilities.
The recent introduction of the ChatGPT software has revitalized the tech industry, which has been hit with numerous job cuts and pledges to reduce ambitious projects. AI has become a hot topic among tech executives, who have been mentioning it much more frequently on earnings calls than in previous quarters.
Read More: Dall-E AI Creates Stunning Humanoid Drawing by ChatGPT
The appeal of AI-driven search is that it can deliver results in simple language, making browsing faster and more effective. However, the impact it may have on targeted advertising remains uncertain.
AI chatbot systems also carry potential risks for corporations, as they can have inherent biases in their algorithms, leading to skewed results, sexualized images or even plagiarism, as seen in previous examples.
Microsoft released a chatbot in 2016 that generated racist content, and an AI used by CNET was found to produce incorrect or plagiarized stories. The Bard ad has been viewed over a million times on Twitter.
