How OpenAI, Google And Anthropic Use AI In Daily Operations

Major AI companies such as OpenAI, Google and Anthropic are already changing how work happens inside their own offices. They are using artificial intelligence in their own workplaces to handle complex tasks and increase productivity.

At OpenAI, the main AI tool used inside the company is called Codex. It was first made for software developers to help them write code, but now even non-technical teams like marketing and recruiting are using it, according to the Wall Street Journal (WSJ) report.

For example, if there is a customer complaint about a billing mistake, earlier an employee would have to contact the billing or operations team to investigate. Codex can do most of that work by itself, such as checking records and finding the issue. The employee then simply reviews and confirms the result.

"This is removing a bunch of bottlenecks and removing reliance on other teams," Ashton Summers, an account director on OpenAI's go-to-market team, told the WSJ.

Codex can automatically create a daily dashboard that shows important customer details and key numbers like how many people are using a product. It even goes through emails and Slack messages and prepares a summary document.

Codex is also doing tasks that usually a junior lawyer would handle. For instance, it can check information shared by new employees and draft replies. It can also flag potential issues, like if someone still has ties to another company or a competitor.

At Google, the company calls itself "customer zero", which means it tests its own AI tools internally before offering them to outside customers. Earlier, employees had to manually check invoices from vendors. Now, Google uses an AI agent that automatically compares invoices with contracts and checks if everything is correct.

"Before, we had a team that actually just mind-numbingly went through the contract. With the new agent, we can review five times more invoices," Van Bui, head of Google Business Services told WSJ.

Google also says this is changing job skills. Workers are learning how to work with AI systems and even describe it as "AI model training" experience on their resumes. 

At Anthropic, employees are using their AI tool Claude to handle routine work. Earlier, a marketing operations employee had to do tasks like creating event pages for campaigns and uploading data manually. One data upload could take anywhere from 15 minutes to an hour. Now, Claude AI does most of this work automatically.

They also use a system where different AI roles work together - one AI "builds" the task, and another AI "checks" or summarizes what was done. The human employee still reviews everything at the end.

Research firm Gartner predicts that in the next two years, the average Fortune 500 company (the biggest companies in the US) could be using more than 1,50,000 AI agents.