Dan Sirk is a so-called fractional executive—that is, he works as a marketing director for not just one company, but two. At the same time.
It’s a juggling act that has become much more manageable with artificial intelligence tools like Claude, Gemini and ChatGPT.
Also read:
Continues after advertising
Previously, it took Sirk three to six months or more to create a custom website with a team of contractors. Now it takes about a month, and he can do it alone.
Developing a communication strategy took a week. When I spoke to him in March, he had just completed this task in less than eight hours. Thanks in part to these efficiency gains, Sirk plans to become chief marketing officer for a third company in the coming months.
Still, when I asked if I should extrapolate recent trends and assume he would add even more companies to his portfolio in the coming years, he looked at me like I was crazy. He insisted that three was the upper limit of what he could manage, even with the help of AI.
“Human relationships still exist,” he protested. Or, more directly: there are meetings.
Sirk estimates that he already attends 10 meetings in a typical week between the two companies. There is a recurring meeting with each senior executive team, not to mention regular one-on-one meetings with each CEO.
There is a meeting with your direct report and another with the head of sales at one of the companies. And there are meetings about specific projects, such as a future presentation for investors from one of the companies.
Continues after advertising
Joining a third company will likely increase meeting volume by 50%. If he became marketing director for one more company, Sirk said, he would spend virtually his entire work week in meetings.
Sirk’s experience, while perhaps extreme, reflects the broader impact of AI on the workplace: It is vastly speeding up many of the tasks performed by office professionals and even replacing some of them entirely.
What it still cannot automate — at least for now — are the structural demands of bureaucracy.
Continues after advertising
With the help of AI, professionals can generate many more memos or strategic options than before and produce more prototypes of products or software features.
But some executive still needs to decide which option to approve. Workers can create many more sales proposals, but they still need to convince customers to close a deal.
As AI makes the production of intellectual work increasingly efficient, the work of presenting, debating, negotiating, lobbying, pressuring, reassuring, or simply selling that work appears to gain in importance.
Continues after advertising
And the need for these sometimes messy human tasks may limit the number of people AI will replace.
“These have always been important skills,” said David Deming, an economist and dean of Harvard College. “But as the information environment becomes more saturated, the ability to tell a story from it — to take a huge amount of text and turn it into something people want — becomes more valuable.”
Can you persuade your colleagues?
The idea that automation increases the importance of in-person interactions is not entirely new. A 2017 study by Deming found that as computers became more powerful, an increasing share of jobs required intense social interaction, while a smaller share required a lot of mathematical knowledge but little social interaction — such as certain engineering roles.
Continues after advertising
By automating technical tasks, computers were, in effect, pushing people into jobs that valued soft skills, Deming noted. This didn’t mean that emotionally skilled people were automatically the most successful—those who did best tended to combine soft skills with technical knowledge—but it changed what employers valued.
In interviews, professionals from a variety of office occupations said that AI has heightened this pattern. Many preferred not to identify themselves for fear of displeasing their employers.
A data scientist at a software company said he and his colleagues used to need to write code for every new feature or improvement they wanted to evaluate. Now they just come up with the idea, and the AI writes the code and performs the analysis.
The company’s selection process, which was previously dominated by questions about programming and favored more introverted profiles, now focuses on assessing whether candidates can identify good ideas and demonstrate the ability to persuade colleagues to support them, he said.
Mark Ozaki, a director at KPMG, said the consultancy had traditionally encouraged younger consultants to specialize in a subject area such as tax law or a technical area such as programming. But AI is devaluing this type of specialization and increasing the value of generalists who take initiative and excel at cultivating customer relationships, he said.
Ozaki, who oversees a team developing an AI-based sustainability platform called Sustainlit.com, said that in the past, his team was sometimes dependent on highly skilled programmers.
Now, it can use AI to do most of the programming and mainly needs people “with the phone glued to their ear, who are friends with everyone, who are always on the move”.
Goodbye, programmers; hello customer success
Cory Crosland, chief executive of PolicyFly, which sells software that helps insurers issue policies, said AI has reduced both the time needed to set up the software for new customers and the number of employees needed to do so.
The move allowed the company to charge much less upfront, which appears to be increasing demand for its services.
To keep up, PolicyFly has grown from 20 to 28 employees in the last six months, and only two of the new hires are software engineers.
Several are younger employees who help onboard customers or work in customer success, helping them extract more value from the software.
Still, Crosland said he doesn’t believe he’ll be able to automate much more of the process, at least in the near future. The reason? Your customers want to interact with a human being.
Customers want PolicyFly to give them confidence that the software will work in different situations and that they have correctly set up billing or are prorating policies appropriately.
And, of course, there are the meetings to sort it all out—many, many meetings. “In larger companies, we have multiple stakeholders weighing in from different departments,” Crosland said. “It’s even harder to reach an agreement and align everything.”
c.2026 The New York Times Company
