The launch of ChatGPT a few months ago kicked off an artificial intelligence arms race among Big Tech companies. Microsoft has invested billions into Open A.I., the company that created ChatGPT, and plans to incorporate the technology into its Bing search engine. And the chatbot’s debut forced Google, which has been researching A.I. for years, to jump into the game and rush to market with its own A.I. product, Bard. 

After decades of search dominance, the rapid developments have left Google playing defense. The company asked its A.I. teams last month to “prioritize working on a response to ChatGPT.” And the urgency for Google to up its game made some staffers lash out about the “rushed” and “botched” announcement of its own chatbot now in beta testing.

In the latest glimpse of the pressure Google is under, newly leaked audio obtained by Insider reveals the company’s Cloud CEO Thomas Kurian hitting back at critics in a company town hall last month, and saying that the battle for A.I. dominance was only just beginning.

“There are other companies taking shots at us saying that we are late to market. We’ve missed the A.I. boat,” Kurian said in the recording. “I would tell you this is the very first minute of a new game, and the game is never done in the first minute.”

Kurian further called on employees to work together, and hyped up the workers laboring over the company’s A.I. plans.

“We are committed as an organization to providing the best capability in these markets, and I can promise you that our engineers are going to deliver amazing technology in A.I. to our customers,” Kurian said. 

Google did not immediately return Fortune’s request for comment.

The company is navigating intense pressure and high stakes in the A.I. race.  Bing, which only controls a fraction of the search market, has much to gain from a successful A.I. search product, while Google has everything to lose. 

“We believe that Microsoft is aiming not only to lower Google’s search margins, but also to dissuade Alphabet from running Google Cloud and other businesses at a loss,” an ARK Invest analyst wrote in a note earlier this month.

While OpenAI is considered a small company that is more able to take risks—like releasing a chatbot to the internet at large—Google must weigh concerns about its reputation and how problematic A.I. released to the public could do damage to the company. 

Its new chatbot, Bard, is currently only available to a limited group of “trusted testers.” And even though Google did take a leap with Bard, the chatbot infamously gave incorrect answers during a videotaped demo announcing the product earlier this month, resulting in a $100 billion wipeout of market value from the California-based tech giant.