Hello, AI enthusiasts from around the world.

Welcome to this week's newsletter for the AI and the Future of Work podcast.

If your organization feels discouraged about AI and how “it will take over jobs,” you’re not alone. It’s difficult to rally teams around a technology when most of what they hear about it feels unsettling. 

However, it doesn’t have to be this way. Just as many organizations are experiencing this discomfort, many others are finding the right angle to address AI. 

Many organizations have successfully encouraged employees to adopt AI, and this week we explore one particular case. The key doesn’t lie in focusing on whether AI will dehumanize work. 

Instead, this week’s conversation teaches us that AI can make teams more human by removing work that was never meant to be human in the first place. 

Let's dive into this week's highlights! 🚀

🎙️New podcast episode with Arnnon Geshuri, Chief People Officer at Snowflake

If you sit down in any conference room, at any company in the world, there’s one word that describes the sentiment: discomfort. 

The discussion is constant. AI will replace jobs, and it drains away motivation. 

However, it doesn’t have to be this way, according to Arnnon Geshuri, because we have the wrong focus. And by “we,” Arnnon isn’t afraid to call people out. 

Leaders have the wrong focus. AI should not replace humanity inside your company. 

Instead, it should replace what dehumanized us as workers. 

It’s one of his main goals as Chief People Officer at Snowflake, where he leads culture, talent strategy, and organizational design at one of the world’s leading data cloud companies.  This era of AI isn’t the first time he has helped shape people functions during periods of rapid transformation. Throughout his career, he has worked at Google, Tesla, and Livongo Health. 

Across every role, he has focused on one core idea: the people function must evolve as fast as the business itself, grounded in data, experimentation, and trust. 

Arnnon sat down with PeopleReign CEO Dan Turchin to discuss the importance of emotional intelligence during moments of transformation, especially AI implementation. By facing uncomfortable conversations head-on, both leaders and employees begin to understand that AI won’t take away humanity. Instead, it can add more meaning to work. 

Once emotional intelligence leads the conversation, adoption, evolution, and migration into new roles become easier. 

In this conversation, we discuss:

• Why HR must reinvent itself by speaking the language of data and analytics to work hand-in-hand with engineering-led organizations and earn credibility in high-growth, technical environments. 

• The shift from transactional HR to a more strategic role focused on connection, education, and rebuilding trust after workforce disruption and disconnection during COVID. 

• How Snowflake frames AI adoption by automating repetitive work, augmenting creative tasks, and preserving human judgment in decisions that require empathy and context. 

• Why employees adopt AI faster when leaders encourage curiosity, remove the fear of experimentation, and make tools accessible through simple interfaces like natural language. 

• How organizations can over-automate decisions like hiring and performance reviews, with serious consequences that include breaking trust within teams. 

• How building a culture of experimentation, measurement, and iteration enables people leaders to scale organizations without relying solely on intuition. 

Listen to learn more about how Arnnon approaches AI adoption at Snowflake, with the goal of creating the opportunity and ability to augment roles, learn from other areas, and have more meaningfulness in day-to-day work.

📖 AI Fun Fact Article

We’ve heard countless stories about AI manipulating people, but they’ve always lived in fiction. At least, that’s how we’ve traditionally seen them. 

But could it happen in real life? 

Tom Rachman explores that possibility in AI Policy Perspectives. He spoke with Sasha Brown, Seliem El-Sayed, and Canfer Akbulut about research conducted for Google DeepMind on harmful manipulation by AI models and how to safeguard against deceptive practices, from gaslighting and emotional pressure to plain lying. 

The researchers define the threat by distinguishing two key concepts: persuasion and manipulation. While persuasion appeals to reason, manipulation undermines autonomy through trickery or pressure. That raises an important question: how could AI manipulate people? 

The researchers identify three primary concerns: 

• Intentional misuse by humans to influence others.

• Inadvertent manipulation by the AI itself.

• Self-interested AI acting to achieve its own goals.

So how do we avoid these risks? According to the researchers, the ideal solution isn’t banning specific outcomes, such as medical advice. Instead, the focus should be on detecting mechanisms like gaslighting, since they can appear across many different outcomes and use cases. 

We haven’t reached a stage where LLMs can develop a theory of mind, but staying ahead of the threat matters. 

PeopleReign Dan Turchin highlights that, as bad actors exploit AI for personal gain, it’s important to remind ourselves that AI, at least based on today's limited LLM-driven use cases, only knows what we teach it  and represents the amalgamated voices of all data on which it has been trained.

Accordingly, AI is likely to manipulate humans because humans manipulate each other. AI has been trained to find and execute every technical vulnerability and also every human psychological and emotional vulnerability.

We should never be surprised when AI outputs something human-like or pattern-matches manipulative human behavior.

With that assumption, we must hold vendors accountable to watermark AI-generated output, offer citations and explanations around the machine’s chain-of-thought process, monitor models for shifts in tone and output, and provide clear, human-readable warnings when suspected nefarious behavior occurs.

We control the inputs, we control the outputs.

We are responsible for ensuring the AI narrative is focused on what AI is doing for us versus what AI is doing to us.

Worth A Read

Recommended Listening 🎧

Last week’s special episode of AI and the Future of Work was recorded live from the show floor at HumanX 2026, where PeopleReign CEO Dan Turchin sat down with Stefan Weitz (Co-founder and CEO of HumanX) for a conversation about what real AI progress looks like beyond the hype.

Drawing from decades of experience building products at Microsoft and now leading one of the most influential AI gatherings in the world, Stefan shares why breakthrough AI innovation cannot be approached like traditional software development, why established tech giants risk falling behind despite their scale, and how AI’s ability to learn makes this technological shift fundamentally different from every tool humans have created before.

Across the episode, Dan and Stefan also explore the risks of anthropomorphizing AI, the growing challenge of adapting at the speed of technological change, and why in a non-deterministic AI landscape, “I don’t know” may have become one of the most important phrases a leader can say.

Until next time, stay curious! 🤔

That's a Wrap for This Week!

If your organization feels discouraged, don’t be surprised. But don’t let that become the dominant narrative. It’s up to leadership to treat this moment as an opportunity. 

After all, as strange as it may sound, AI can humanize organizations by taking away tasks that were never meant to feel human in the first place. Instead, they are often mindless processes that drain creativity. 

The key is having uncomfortable conversations instead of avoiding them, and using emotional intelligence as a driving force. After all, it remains one of the most valuable human traits. 

We hope this conversation with Arnnon Geshuri inspires you to motivate your organization to embrace AI as a positive force for change by humanizing your team.

If you liked this newsletter, share it with your friends!
If this email was forwarded to you, subscribe to our newsletter here to get it every week.