Audio Sources - Full Text Articles

Google execs say the company isn’t launching a ChatGPT competitor because it has greater ‘reputational risk’ than startups like OpenAI

Alphabet CEO Sundar PichaiAlphabet CEO Sundar Pichai.

Jerod Harris/Getty Images

  • Google executives reportedly told employees they don’t think chatbots are ready to replace search.
  • Employees asked execs if failing to launch a chatbot like OpenAI’s ChatGPT is a “missed opportunity,” CNBC reported. 
  • Alphabet’s CEO and Google’s head of AI said the company has a greater “reputational risk” than startups like OpenAI.

Google executives reportedly told employees that the company does not plan to launch a chatbot to compete with OpenAI’s buzzy ChatGPT because it’s a “reputational risk.”

Speaking during a recent all-hands meeting, Google’s Head of AI Jeff Dean said despite having artificial intelligence technology products and capabilities, the company has to make decisions “more conservatively than a small startup,” CNBC reported.

His remarks came in response to a question from an employee about whether failing to make a rival chatbot would be a “missed opportunity,” particularly given that Google has its own conversation technology such as LaMDA, or Language Model for Dialogue Applications. 

Dean, who was joined in the meeting alongside Alphabet CEO Sundar Pichai, said that while Google’s technology is just as capable as the trending chatbot, the company is more vulnerable to chatbot-related issues like bias and false information because over a billion people use and depend on Google to search for information.

“We are absolutely looking to get these things out into real products and into things that are more prominently featuring the language model rather than under the covers, which is where we’ve been using them to date,” Dean said about Google’s AI. “But, it’s super important we get this right.”

Other concerns include the limitations of bots like ChatGPT listed in a Nov. 30 OpenAI blog post, including “plausible-sounding but incorrect or nonsensical answers,” “harmful instructions,” and “biased behavior” in some of its answers. 

OpenAI CEO Sam Altman has also acknowledged ChatGPT’s limitations, tweeting on Dec. 10: “ChatGPT is incredibly limited, but good enough at some things to create a misleading impression of greatness. it’s a mistake to be relying on it for anything important right now. it’s a preview of progress; we have lots of work to do on robustness and truthfulness.”

ChatGPT reached over one million users within five days of its launch, and is backed by Google’s rival, Microsoft. OpenAI was co-founded by Sam Altman and Elon Musk in 2015. Musk resigned from the company’s board of directors three years later.

Google did not immediately respond to Insider’s request for comment.

In addition to LaMDA, Google’s BERT and MUM AI language models, which are used to improve its search engine, are competitive with ChatGPT. MUM, which stands for Multitask Unified Model, can understand information from different mediums, like from a webpage and from a photo, at the same time, to give users an answer.

A former Google executive told Bloomberg that ChatGPT could “disrupt” Google’s ad business by stopping users from clicking links that have ads. Sridhar Ramaswamy, who was the head of Google’s ad team from 2013 and 2018, told Bloomberg that ChatGPT was “a better experience” for searches.

But a Morgan Stanley report published on Monday said the firm thinks ChatGPT will not be a large threat to Google’s position as it continues improving its search engine and language models.

In the all-hands meeting, Pichai reportedly said Google has “a lot” of plans for AI in 2023.

Read the original article on Business Insider