The human cost of AI is something we don’t talk about nearly enough. In this episode, I sat down with Sharon Nannetti, Director of Global Services at Zevo Health, to explore the hidden world of trust and safety, content moderation, AI red teaming, and the very real psychological toll on the people keeping our digital world safe.
Who Is Sharon Nannetti?
Sharon describes herself as an innovator and a rebel — and she means it. With a background in digital transformation, psychotherapy, and neuroscience, she spent much of her career in Australia before returning to Ireland. She now leads global services at Zevo Health, an organisation that provides psychological support to high-pressure teams including trust and safety professionals, AI red teamers, fraud teams, and call centres.
What Is Trust and Safety?
Trust and safety is an ecosystem — corporations, platforms, regulators, and government bodies all working together to protect users from fraud, abuse, and harm on digital platforms. But inside that ecosystem are hundreds of thousands of human moderators who review content every single day so the rest of us don’t have to.
The Hidden World of AI Red Teaming
Red teaming is the evolution of content moderation. Red teamers spend eight hours a day thinking like the worst actors imaginable — generating extremist content, exploitation material, and violent prompts — all to find the gaps in AI guardrails before bad actors do. Sharon called them the heroes of the world, and honestly, after this conversation, I couldn’t agree more.
The psychological effects Sharon sees most in red teamers include:
- Moral injury — constantly operating against your own values
- Identity strain — blurring the line between the role and the real person
- Creative fatigue — the relentless pressure to generate new and unique prompts
- Burnout — the cumulative result of one or all of the above
Sharon shared the story of a red teamer named Luigi who stopped bringing his children to the playground — not because he posed any threat, but because months of working on child sexual abuse material had made him distrust the psychological effect the work was having on him. That stopped me in my tracks.
The Performance Paradox
Here’s something that really hit home for me. AI was supposed to give us back time and reduce cognitive load. But what’s actually happening is that by removing simple tasks, AI is leaving humans with only the complex, high-stakes work — with no recovery time in between. Sharon pointed to an NHS call centre in the UK where agents now handle all simple calls, and burnout rates among the human staff have gone through the roof.
Sharon calls this the performance paradox: the very parts of our brain needed for creativity, judgment, and good decision-making are the parts we deplete through chronic stress. And AI, right now, is accelerating that stress.
What Leaders Need to Do
Sharon’s advice for leaders is clear — well-being can’t be a yoga class on a Wednesday. It has to come from the top, embedded into how work is designed, how workloads are structured, and how recovery is built in. She put it simply: your people need to work in partnership with AI, not just as users of it. And that requires the right environment and conditions to make that possible.
Where Are We Headed?
Sharon’s team recently did strategic research on where the trust and safety industry would be by 2031. When they presented at the global Trust and Safety Conference in San Francisco, they realised what they thought would take five years will likely happen in about eighteen months. The pace is faster than any of us have realised.
If you want to build your AI literacy and skills, I can help
📚 Grab AI Playbook on Amazon, the perfect podcast companion.
📰 Subscribe to Human in the Loop, my LinkedIn Newsletter
🎓 Take the CPD Accredited AI Playbook Diploma
📞 Book a call at digitaltraining.ie
Key Statistics
– There are hundreds of thousands of content moderators around the world reviewing digital content daily.
– Red teamers spend eight hours a day generating adversarial content to test AI guardrails.
– Zevo Health has been working with trust and safety teams for approximately ten years.
– Mark Zuckerberg published a six-and-a-half-thousand word essay on AI and agentic assistants.
– What Sharon’s team projected would take until 2031 in trust and safety is now expected to happen within eighteen months.
– An NHS borough in the UK introduced AI agents for call centres, resulting in off-the-scale burnout among human staff handling only complex calls.
About the Podcast Host
AI and Communications Specialist | Author | Podcaster | International Speaker
Joanne Sweeney is a recognised leader in digital marketing and digital transformation, specialising in the practical application of AI to enhance communication strategies. The founder of Digital Training Institute, Joanne, empowers leaders and communications professionals to leverage AI responsibly and effectively, ensuring transparency, trust, and tangible results.
She completed the Harvard Kennedy School Executive Programme in AI and Public Policy this summer before launching her fourth book AI Playbook, in September. Joanne has also published two editions of Public Sector Marketing Pro, which were preceded by her first title, Social Media Under Investigation Law Enforcement and the Social Web.
Her books have become go-to resources for governments, public sector and policing agencies, as well as marketing and communications professionals, all navigating the challenges of the changing Internet.
Joanne is the host of the AI SIX Podcast, a daily six-minute podcast that provides actionable insights, expert interviews, and practical advice to help professionals navigate the new world of work.
