AI Future of Work: Hassan Sawaf on UNFLTRD with Muhammad Chbib

The global conversation around artificial intelligence (AI) often swings between two extremes: fear of mass job loss or hype about limitless potential. In his recent appearance on UNFLTRD with Muhammad Chbib, our CEO Hassan Sawaf offered a more grounded perspective: AI is not replacing us. It is shaping the AI future of work, governance, and innovation.

The conversation rejects two lazy extremes: AI ends work and AI solves everything. Instead, it argues for something harder and better: designing AI to amplify the 99% through real-time governance, agent societies that explain themselves, and institutions that reward experimentation over perfection. Ultimately, the upside is massive; the responsibility is ours.

This framing set the stage for Hassan’s insights on the AI future of work.

With a career spanning eBay, Amazon, Meta, and now aiXplain, Hassan brings rare depth: moving beyond theory into lived experience at the forefront of AI.

Here are four key takeaways from the conversation.

1. Jobs Will Evolve, Not Disappear

When Chbib cited the IMF’s projection that 40% of jobs worldwide are at risk, Sawaf countered:

“Jobs aren’t disappearing. They’re evolving with AI.”

For example, Hassan recalled how machine translation at eBay didn’t eliminate translators but instead created around 150,000 new jobs through cross-border trade.

Consequently, AI creates opportunity by enabling humans to focus on higher-value tasks while agents take on repetitive work.

2. Governance Is the Difference Between Innovation and Chaos

In response to Muhammad’s questions on AI oversight, Sawaf explained that simply having guardrails is not enough. Therefore, enterprises need Inspectors, Bodyguards, and kill switches built into their agent architectures. Similarly, just as every society requires rules and enforcement, every AI ecosystem requires governance to ensure safe adoption.

“AI without governance is not innovation. It is chaos.”

Guardrails behave like laws, useful but breakable. Enterprise systems need active controls:

• Inspector-style agents that check every step for truthfulness, safety, policy alignment, and required disclosures.
• Bodyguard-style agents that enforce data minimization, redaction, jurisdiction rules, and consent.
• A breaker or kill switch at the runtime level to stop misbehavior, plus traceability to explain why decisions were made.

In this way, governance shifts from after-the-fact audits to real-time enforcement. That is the only tempo that matches the technology.

As a result, governance turns AI from risk into a trusted enterprise tool.

3. Production-Ready Agents Are the Future

Hassan drew a line between demos and systems built for the enterprise:

“The difference between demos and production is governance, resilience, and compliance.”

For enterprises, success requires agents that are auditable, resilient, and compliant. In addition, in industries such as healthcare and finance, this means meeting regulatory requirements, self-correcting when errors occur, and providing transparency in decisions.

Ultimately, enterprises need production-ready agents, not experiments, to deploy AI at scale.

4. The Middle East Has a Chance to Lead

When Chbib asked about the Middle East’s role in AI, Sawaf explained that, historically, the region has been a consumer of global technology. However, he now sees this trend changing.

“The Middle East must move from being a consumer of AI to becoming a creator.”

As a result, with investments in compute, large language models, and innovation ecosystems, the region can shape the future of AI: building systems that reflect its strengths and values.

Finally, by moving from consumer to creator, the Middle East can reclaim its role as a driver of innovation.

What This Means for the Future

AI is no longer a distant future. It is here, reshaping work, governance, and society. Sawaf stressed that the real question is not whether AI will change our world, but how we guide that change. At aiXplain, we believe the answer lies in an Agentic OS: production-ready agents, active governance, and the tools to build AI systems that scale responsibly.

Explore how the Agentic OS can help your enterprise scale AI with confidence