Compiled by the editorial desk with reference to official statements, public filings, and industry data.

In a move that has raised eyebrows across the tech industry, Google released its Bard AI chatbot to the public last month despite a torrent of internal criticism from its own employees. According to a Bloomberg report, approximately 80,000 staff members were enlisted to test the unreleased chatbot, and their feedback was overwhelmingly negative, with one employee describing the AI as a “pathological liar” and another labeling its responses “cringe-worthy.”

The internal dissent was not limited to casual observations. A February note, seen by nearly 7,000 employees, went so far as to call Bard “worse than useless: please do not launch.” That note, along with other critical feedback, was obtained by Bloomberg, painting a picture of a product that many inside the company believed was not ready for public consumption.

Despite these warnings, Google proceeded with the launch, framing Bard as an “experiment” and attaching prominent disclaimers to its outputs. The decision appears to be a strategic response to the rapid rise of OpenAI's ChatGPT, which has captured significant public attention and market share. Analysts suggest Google felt pressured to field a competing product, even if it meant releasing a tool that many employees considered underdeveloped.

Ethics Concerns Take a Back Seat

The episode has also reignited debate over Google's commitment to AI ethics. Bloomberg reports that employees tasked with evaluating the safety and ethical implications of new products were instructed to stand aside during the development of AI tools. Meredith Whittaker, a former Google manager and president of the Signal Foundation, told Bloomberg that “AI ethics has taken a back seat.” She added, “If ethics aren’t positioned to take precedence over profit and growth, they will not ultimately work.”

Google maintains that AI ethics remain a top priority, but two employees told Bloomberg that ethical reviews are largely voluntary, except in sensitive areas like biometrics or identity features. The company's own executives have signaled a willingness to relax standards. During a recent meeting, Google executive Jen Gennai suggested that upcoming AI products could score an “80, 85 percent, or something” on a fairness metric, rather than aiming for a higher threshold of 99 percent.

The internal friction has led to frustration among those tasked with ensuring responsible AI development. Margaret Mitchell, who led Google's AI ethics team until her departure in 2021 following a fairness dispute, described the current atmosphere as one of deep disillusionment. “There is a great amount of frustration, a great amount of this sense of like, what are we even doing?” she told Bloomberg. “Even if there aren’t firm directives at Google to stop doing ethical work, the atmosphere is one where people who are doing the kind of work feel really unsupported and ultimately will probably do less good work because of it.”

Google's decision to launch Bard despite the internal backlash highlights the growing tension between innovation and responsibility in the AI sector. As the race to dominate the market intensifies, the company's handling of Bard may serve as a case study in the perils of prioritizing speed over safety.