While many assume formal learning is limited to a bachelor’s or master’s degree, both the CEO of VC firm General Catalyst, Hemant Taneja, and McKinsey’s top executive, Bob Sternfels, say that’s not the case anymore.

Employees must skill and re-skill constantly to stay afloat, said Taneja, whose VC firm has invested in companies such as Anduril and Anthropic. Taneja discussed this during  a live taping of the All-In podcast, hosted by entrepreneur and investor Jason Calacanis Tuesday at CES 2026, a massive annual tech trade show in Las Vegas.

“This idea that we spend 22 years learning and then 40 years working is broken,” Taneja said.

Yet, in a workplace where AI agents can be trained quicker than employees, workers don’t only need knowledge, they must find ways to stay relevant, said host Calacanis, who himself made early investments in trading app Robinhood and Uber.

“You’re going to have to show chutzpah, drive, passion,” he said.

McKinsey’s global managing partner, Sternfels, said he’s seen first-hand how AI is transforming the workplace. McKinsey has used the tech to grow client-facing consultant roles by 25%, simultaneously cut the same number of jobs in non-client-facing roles, all the while increasing its output 10% overall.

McKinsey will have just as many AI agents as it has human employees by the end of the year, Sternfels said. Currently, its human employees outnumber AI agents 40,000 to 25,000.

“Our model has always been synonymous that growth only occurs with total head count growth. Now it’s actually splitting,” Sternfels said. “We can grow in this part, the client-facing side, and we can shrink in this part and have aggregate growth in total.” 

AI job anxiety

McKinsey’s stark results from incorporating AI agents play to the heart of workers’ fear about how AI will disrupt their jobs as adoption increases. Some could argue young workers have a right to worry. 

A study by the Stanford University Digital Economy Lab in November found early career workers between the ages of 22 to 25 in occupations at most risk of disruption have experienced a 13% relative decline in employment levels since 2022, when OpenAI released ChatGPT. Another study by Gallup from last year found 37% of workers claimed their workplace had implemented AI.

As Sternfels and Taneja said, the added pressure of AI means learning and evolving is more essential than ever. Yet, some have pushed back on the idea that more AI means entry-level workers don’t matter.

Despite the increased push by Amazon CEO Andy Jassy to implement AI at the company, Matt Garman, the CEO of subsidiary Amazon Web Services, has repeatedly said entry-level workers are essential to a healthy organization. Without entry-level workers, “you have no talent pipeline,” Garman told Wired.

Thus, replacing entry-level workers with AI is “one of the dumbest things I’ve ever heard,” Garman said on the Matthew Berman podcast last year.

Now OpenAI is leaning into this trend as it seeks to build out more products to keep users engaged and establish itself as the new digital interface that will mediate interactions between different industry sectors—from e-commerce to finance to, yes, wellness and healthcare—and their customers.

Today OpenAI announced the debut of ChatGPT Health — a dedicated experience inside ChatGPT where it says users can securely connect medical records and wellness apps such as Apple Health, Function, and MyFitnessPal to further personalize conversations. OpenAI said it would not train its models on personal medical data.

During a press preview, Fidji Simo, OpenAI’s CEO of applications, introduced ChatGPT Health by sharing a personal story about how ChatGPT helped her after she was hospitalized for a kidney stone last year and developed an infection. A resident had prescribed a standard antibiotic, but she checked it against her medical history in ChatGPT, which flagged that the medication could reactivate a serious life-threatening infection she had suffered years before.

“The resident was relieved I spoke up, she told me she only has a few minutes per patient during rounds, and that health records aren’t organized in a way that makes it easy to see,” she said. “I’ve heard many stories like this from people who are using AI to help connect the dots in their healthcare system that really wasn’t built to see the full picture.”

Five months ago, OpenAI signalled its push into healthcare with two high-profile hires: Nate Gross, the cofounder and former chief strategy officer of Doximity, a leading digital platform for medical professionals, leads OpenAI’s healthcare strategy. Ashley Alexander, former co-head of product at Instagram, leads healthcare product. But Karan Singhal, who leads health AI at OpenAI, said during the press preview that the company had been laying the groundwork for ChatGPT Health for about two years.

According to an OpenAI blog post, the company analyzed deidentified ChatGPT conversations and found that more than 230 million people globally ask health- and wellness-related questions on ChatGPT every week. Even with that massive user base, however, that doesn’t mean the company has an easy road ahead: They are joining a fast-moving race among Big Tech and startups to become the AI front door for consumer healthcare. 

For example, for ChatGPT Health OpenAI has partnered with b.well, a health management platform which combines patients’ health records, financial information, and wearable and other healthcare data. But Google also announced a partnership with b.well in October 2025 – potentially setting the stage for future AI-driven consumer health tools, though the internet giant has not yet announced a health-specific feature set for its AI chatbot product Gemini. 

ChatGPT Health will not be widely available immediately. There is a waitlist for access to a small group of early users, but the company said it will make Health available to all users on web and iOS in the coming weeks. Electronic Health Records (EHR) integrations and some apps are available in the U.S. only. 

OpenAI does not describe ChatGPT Health as HIPAA compliant (consumer health apps are not covered by HIPAA) but says it adds layered protections for sensitive health data and excludes health conversations from model training by default. It also says users can enable multi-factor authentication (MFA) to add an extra layer of protection to help prevent unauthorized access, and that access to medical records can be removed at any time in the “Apps” section of Settings.”

ChatGPT Health also fits a broader pattern at OpenAI: building vertical-specific experiences on top of its core models, rather than relying solely on general-purpose chat. In education, the company launched Study Mode, a ChatGPT spinoff for students, in July to compete with Google’s Gemini for Education. It has also rolled out agentic shopping and shopping research features. There are also reports that a finance-specific experience is also in the works.

OpenAI also clarified that ChatGPT Health was not part of the company’s eight-week long “code red,” announced in early December by Sam Altman in an internal memo. “[ChatGPT Health is] actually sitting outside of Code Red,” said Simo. “We have been working on health for a very long time…we know this is a core use case of ChatGPT, we’ve known that we wanted to make that use case even better and serve even more of people’s needs with the changes that we’ve made.”