Join top executives in San Francisco on July 11-12, to hear how leaders are integrating and optimizing AI investments for success. Learn More
At the beginning of last week, the hype around generative AI seemed to hit Super Bowl-level intensity.
But rather than the Philadelphia Eagles facing off against the Kansas City Chiefs, it was Google’s Bard competing against Microsoft’s Bing in a pair of generative AI launch debuts that many hoped would show off a fierce competition between two tech titans — leaving audiences spellbound with the new possibilities for web search.
>>Follow VentureBeat’s ongoing generative AI coverage<<
Unfortunately, the highly-anticipated matchup didn’t end up matching up to expectations.
Join us in San Francisco on July 11-12, where top executives will share how they have integrated and optimized AI investments for success and avoided common pitfalls.
Instead, Microsoft — along with OpenAI, whose ChatGPT-like model powers the new Bing — clearly won the PR victory, with clear and powerful verbiage from CEO Satya Nadella like “The race starts today” in search.
Google, on the other hand, delivered a muted performance that included unforced errors like a mobile phone gone missing during a demo and, far more costly, an error in an ad touting Bard’s talents — powered by LaMDA — that cost the company over $100 billion in stock value.
Both Google’s and Microsoft/OpenAI’s chatbots hallucinate
Large language models (LLMs), like OpenAI’s ChatGPT and Google’s LaMDA, undoubtedly have the power to change how we search for and find information. But was Google’s failed Bard launch — which Google employees criticized for being “rushed and botched” — really that bad?
After all, Microsoft’s new Bing, which is based on a more powerful version of ChatGPT and one customizable for search, has the exact same problems as Bard (and every other LLM for that matter): Hallucinations, or confident-yet-made-up answers.
Yet, Microsoft’s launch was hailed, while Google’s was panned.
AI critic Gary Marcus, in a blog post last week, said he would always remember February 8, 2023, as the day in which a chatbot-induced hallucination cost Alphabet $100 billion.
“But I will also remember it as the week in which Microsoft introduced an ostensibly similar technology, with ostensibly similar problems, to an entirely different response.”
No one has explained, he continued, why Google and Microsoft received such different receptions.
“The two mega-companies both demoed prototypes, neither fully ready for public use, built around apparently comparable technology, facing apparently similar bugs, within a day of each other,” he wrote. “Yet one demo was presented as a revolution, the other as a disaster.”
The search for a reliable AI chatbot for search
Prabhakar Raghavan, senior vice president at Google and head of Google Search, warned Germany’s Welt am Sonntag newspaper about in an interview published on Saturday about chatbot hallucinations.
Raghavan said Google is still conducting user testing on Bard and has not yet indicated when the app could go public.
“We obviously feel the urgency, but we also feel the great responsibility,” Raghavan said. “We certainly don’t want to mislead the public.”
Google’s Big Game loss was certainly bad for shareholders. But however the generative AI search battle plays out, larger questions around AI chatbot reliability will need to be addressed if widespread consumer adoption is to be expected and encouraged. If not, as Gary Marcus points out, people may tire of trying to decipher truth vs. BS in their everyday searches.
Honestly, after a week of Super Bowl-level AI hype, I’m ready for some non-hallucinating, straight-up facts. For that, maybe I’ll just old-school Google it.
VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact. Discover our Briefings.