How smart companies are keeping workers on side as AI reshapes their jobs
More than half of Americans fear AI will cost someone in their home a job. A handful of companies are trying a different approach, and it starts with telling the truth.
Key points
- 53% of Americans across demographic groups worry that AI will put someone in their household out of work, according to recent survey data.
- A 2026 Software Finder report found 53% of workers feel AI tools make their role seem less necessary.
- Flavoured syrup maker Torani, which has recorded zero layoffs across 103 years, plans to grow its workforce by 30% by March 2027, backed by a $60 million manufacturing expansion.
- Gartner managing partner Jackie Swanson says almost no organisation has an honest plan for what AI is doing to its people, even when it has a full technology rollout roadmap.
- Contract software firm Ironclad and productivity company Superhuman each frame AI as a skill upgrade rather than a cost-cutting exercise.
Workers are nervous. That is the plain truth sitting underneath every corporate AI announcement right now. More than half of Americans worry AI will put someone in their household out of work, and that fear has shown up in election debates and, reportedly, in investor documents for some high-profile company stock listings.
A lot of that anxiety, says Jackie Swanson, managing partner at research firm Gartner, comes down to a communication gap. Every company has a plan for rolling out AI tools. Almost none of them have a straight answer for what those tools mean for the people doing the work.
What are companies actually getting wrong?
The problem is not the technology. It is the silence around it. When workers hear "efficiency" and "automation", their brain fills in the blank with "redundancy". The companies having the most success are the ones replacing that blank with something specific and honest.
Take Ironclad, a contract software firm founded in 2014. Its chief technology officer, Sunita Verma (previously at Google and Character.AI, the company behind popular AI chatbot apps), frames the whole thing as acceleration rather than cost-cutting. Her argument: efficiency has a ceiling, growth does not. More practically, her team ran internal training sessions, encouraged peer learning, and was upfront about where the tools fall short. Coming from someone who actually understands the technology, that candour lands differently than a memo from HR.
Grammarly, the writing assistant tool most people have used at some point, rebranded as Superhuman in 2025 after acquiring an AI email app of the same name and a productivity tool called Coda. It is now firmly an AI-first company. Chief people officer Kenny Mendes takes a deliberately bottom-up approach to change. "We don't start with a top-down AI mandate," he told CNBC Tech. Teams identify their own pain points and pick their own tools. The result, he says, is that people feel like they are driving the change rather than having it done to them.
Can a no-layoffs track record actually help?
Yes, and Torani is the proof. The California syrup manufacturer has not laid anyone off in 103 years. That history means workers already trust the organisation when it says a new technology is not a threat.
CEO Melanie Dulbecco has run the company since 1991 and says she never uses the word "acceleration" with staff. Instead the question she asks is: how do we make work better, less manual, more interesting? Torani has a formal commitment that it "will not eliminate work through technology", and its 30% workforce expansion by March 2027 backs that up in concrete terms.
The lesson across all three companies is similar. Workers do not need a perfect answer. They need to trust the person giving them one.
What does this mean for ordinary employees?
If your company is rolling out AI tools, the single most useful thing you can do is ask a direct question: what is the plan for my role? A confident, specific answer is a good sign. Vague talk of "transformation" is worth following up on.
And if you are the person doing the communicating, Dulbecco puts it best: "People need to trust that what they're hearing has a context that they believe." No communication strategy survives a leadership team that does not actually mean what it says.
