In 2024, it was announced that HumanX had raised $6million to launch a premier AI community event which just took place at the Fontainebleau Las Vegas, from 9 to 12 March 2025.
Started by the team behind HLTH and Money20/20, there was understandably a lot of curiosity about what a premium AI event could look like. With senior names from tech companies, finance institutions and e-commerce alike, the event aimed to distill AI complexity into practical and actionable solutions for organisations.
“We’re not here to feed the hype or create more fear,” said co-founder Stefan Weitz. “The conversation around AI is too binary, either we’re talking benevolent robot overlords or Skynet. But the truth is more complicated. We want to talk about what’s really working and what isn’t, and, frankly, where we go next.”
And while that might sound refreshing, it’s clear that AI, even in its more grounded discussions, still has a long way to go in terms of practical, universally applicable solutions but with stellar names from the world of tech, finance and retail – the event certainly tried to get difficult conversations out into the open.
The good, the bad and the Agentic AI
One of the hot topics was Agentic AI, a term that seemed to win the unofficial ‘buzzword bingo’ of the event. It refers to AI systems that can autonomously make decisions and carry out tasks without much human input. According to Adam Evans, EVP & GM at Salesforce, the company’s AgentForce platform, which automates customer service, has already been adopted by over 5,000 customers in just a quarter.
But let’s be honest, there’s a bit of a hard-sell vibe when companies start talking about replacing human workers with AI. Even though the stats look good, you can’t help but wonder if the pitch is more about the future than the present. Sure, automation can make businesses run more smoothly, but the idea of AI running entire departments raises more than a few eyebrows.
If AI systems could fully replace customer service agents, will they really uphold the same level of brand personality and customer care? And if they can, will customers actually accept that?
“We’re seeing a major shift towards automation in service sectors,” Evans said. “And in some ways, these tools are already enhancing productivity. But the question is: How much do we trust these agents to actually represent a brand in the right way?”
For a conference that touted practicality, there was a noticeable disconnect between the optimistic vision of AI as a productivity booster and the real-world implications of this technology for workers. Maybe it’s just a growing pain in the AI revolution, but it’s clear that companies are trying to sell us on the future while we’re still grappling with today’s AI limitations.
Kamala Harris weighs in: innovation vs. trust
Former Vice President Kamala Harris was one of the big draws at the event, offering her perspective on AI’s relationship with public policy. In her conversation with Feedzai CEO Nuno Sebastião, Harris called out the tension between Silicon Valley and Washington, pointing to the erosion of trust between the tech industry and the government. “Right now, that’s a really bad relationship,” Harris said, noting how the last decade has weakened support for research and innovation at the government level.
“We need to figure out how to make this work,” Harris continued. “Because if we don’t, we are losing this very specific moment in time.”
Harris’s comments underscored a critical issue that continues to be a point of contention in AI’s growth trust. As AI becomes more integrated into society, concerns about job loss, misinformation, and AI’s impact on vulnerable communities only continue to grow. She advocated for a more transparent approach to AI development, emphasising collaboration between government and industry to navigate these challenges.
Yet, as much as her conversation was a call for proactive collaboration, it also highlighted the limitations of AI governance as it stands today. The elephant in the room…Whether these lofty ideas will ever translate into real, effective policy and regulatory action before AI takes off in a way we can’t control.
The financial sector: underneath the hood
The financial services sector was well-represented at HumanX, and for good reason. AI has the potential to revolutionise everything from fraud detection to risk management.
Interesting panel sessions included Balancing Risk and ROI: The Role of AI in Financial Services which brought together key industry leaders including Anthony Soohoo, CEO of MoneyGram, Terah Lyons, MD global head of AI and data policy at J.P. Morgan Chase, and Alexander Statnikov, co-founder and CEO at Crosswire Risk Management. The session, moderated by Jared S. Taylor, founder of Slice of Media, addressed AI’s potential to improve compliance and manage risk while also increasing operational efficiencies.
Lyons provided insight into J.P. Morgan’s approach, explaining how AI has been instrumental in improving fraud detection by reducing false positives and enhancing overall operational efficiency. Meanwhile, Statnikov forecasted that by 2025, AI would be central to compliance within financial services, especially as regulatory complexities continue to grow. He warned that financial institutions failing to embrace AI risk being left behind in a rapidly evolving landscape.
Soohoo insights
In an exclusive interview following the panel session Soohoo described his first four and a half months at MoneyGram as a ‘re-founding’ period reimagining the company’s future with AI as a core pillar. For Soohoo, the challenge wasn’t just about adopting new technologies but making sure they aligned with a clear strategy. He stressed that AI shouldn’t just be layered onto existing systems but should be used to first identify and correct inefficiencies before being integrated.
“What I see right now across the space, and that’s not a critique as much as an observation, is that I see a lot of people running towards AI, but without an actual plan,” said Soohoo. AI may be a game-changer, but it’s not a magic bullet especially for large, complex organisations like MoneyGram.
Noting that MoneyGram’s global payment network is one of the company’s most valuable assets, he added that AI could play a pivotal role in expanding its use case particularly for business-to-business transactions and potentially even as a crypto on-ramp. These emerging use cases hint at AI’s transformative potential in the financial sector, but Soohoo was careful to avoid overhyping its benefits.
“We’re still figuring out how to build a solid foundation for AI to enhance what we do,” he added. “We’re talking about system architecture before we get to AI. If your organisation isn’t streamlined, AI is only going to amplify inefficiencies.” Highlighting that the true potential of AI will be realised when it is fully integrated into the system, and why it’s crucial to streamline operations before automating them. A difficulty for legacy tech.
Building trust into AI
A key panel in the Finance track was How AI is Reshaping Risk and Trust in Business featuring Prem Natarajan, chief scientist and head of enterprise AI at Capital One, Laura Spiekerman, co-founder at Alloy, and Rajat Taneja, president of technology at Visa. Moderated by Hugh Son, banking reporter at CNBC, the discussion focused on AI’s ability to build trust in financial services, especially in areas like fraud detection and customer interactions.
The panel touched on some interesting concepts, such as integrating human oversight into generative AI tools to keep customer relationships intact. Natarajan discussed how “observability” and telemetry essentially, tracking and understanding customer behaviour are crucial to building trust with AI systems.
This kind of nuanced approach is what many see as the future of AI in financial services. “AI tools are powerful, but they need to work hand-in-hand with human judgment to ensure we’re meeting our customers’ needs,” he explained.
Taneja provided insights into how AI has revolutionised fraud detection at Visa, enabling faster model development to combat emerging threats like synthetic fraud. He stressed the critical role AI plays in strengthening defences against increasingly sophisticated fraud tactics.
While Spiekerman advocated for AI-driven models that allow businesses to manage fraud risks without the long delays associated with manual processes. As AI continues to reshape fraud prevention, Spiekerman emphasised the need for better integration of models to reduce false positives and improve efficiency.
Agentic AI again
The panel also touched on the growing concept of Agentic AI, which can autonomously predict and perform tasks within a financial enterprise context. This shift, according to experts, will significantly change the way businesses handle risk and decision-making, and it’s a trend that the finance industry will need to keep an eye on.
Visa later unveiled new research in collaboration with the Institute for the Future, exploring the impact of AI on commerce. The study highlights four transformative trends: the expansion of personal identity across smart devices requiring stronger trust networks, the evolving role of AI assistants in reshaping buying and selling behaviours, Generation Alpha’s unique approach to money influenced by gaming and AI tools, and the emergence of markets focused on sustainability as resources become more quantifiable. Visa is set to showcase further advancements in AI and the future of commerce next month in San Francisco.
As the finance world grapples with growing regulatory demands and an increasing need for real-time decision-making, AI presents both a solution and a challenge. The key takeaway from these panels was clear: to stay competitive, financial institutions must adopt AI thoughtfully, ensuring it works in concert with human intelligence and aligns with regulatory frameworks.
Fighting crime with AI
Also launched at the event was a new TRUST Framework established by Feedzai for responsible generative AI. This five-pillar approach: Transparent, Robust, Unbiased, Safe & Secure, and Tested, provides a comprehensive blueprint for developing ethical, secure, and high-performing AI solutions.
Designed to help B2B organisations innovate responsibly, the framework emphasises the importance of building fairness, explainability and security into AI systems to ensure compliance and protect reputations. The initiative aims to differentiate companies in a commoditised AI market by making ethical practices a competitive advantage.
Speaking to The Fintech Times, Andy Renshaw, SVP of product management at Feedzai, was bullish about AI adoption throughout financial services. “This is the year of AI orchestration,” he said, highlighting how AI could now automate everything from fraud detection models to customer communications. “We’re moving from just having AI or not to actually orchestrating AI in ways that help businesses stay on top of fraud without constantly reinventing the wheel,” he explained.
The team used the event to showcase a new fraud prevention product, ScamAlert, which uses AI to gamify a consumer’s role in detecting scams. Feedzai’s game starring Fido was also in demand at the show, as delegates competed for a leaderboard place by spotting skills through emails and texts in timed conditions against the agent (spoiler alert, it was faster and more accurate than a human being every time).
What’s next for HumanX?
It would be nice to wrap up this article with a ‘see you next year in Vegas’ but the truth is that with so many big tech companies that attended it’s no surprise that the 2026 HumanX event will take place in San Francisco. A win for the environment potentially but will it have the same frantic energy and the ‘let’s get things done’ attitude as in Vegas? That’s to be decided.
One thing is for sure, HumanX has set a high standard for AI focused events, from big name speakers to luxury food and drink options in an opulent setting. It also felt – dare I say it – fun.
The AI generated images of the speakers that welcomed you into the event felt modern, the dog park was a huge hit and the dedicated speaker roundtables allowed delegates to get close to the action. There will be large expectations for next year, certainly.
