Beyond the Hype: How Unlikely AI is Tackling Enterprise AI’s Trust Deficit

The Trust Barrier in Enterprise AI Adoption

While artificial intelligence promises to revolutionize business operations, a significant trust gap prevents widespread enterprise adoption. William Tunstall-Pedoe, the mastermind behind Amazon’s Alexa, understands this challenge intimately. His current venture, Unlikely AI, aims to solve what he identifies as the fundamental barrier: reliability in AI systems., according to industry reports

The Trust Barrier in Enterprise AI Adoption
The Unlikely AI Solution: Accuracy, Explanation, Consistency
The High Stakes of AI Implementation
From Alexa to Unlikely: Building Trustworthy AI
The UK’s AI Ecosystem and Global Competition
The Future of Enterprise AI Adoption

“Trust is a theme that is coming up all the time in the industry with artificial intelligence and the major reason why AI isn’t being adopted by businesses,” Tunstall-Pedoe explains during our conversation at The Black Penny in London’s Covent Garden. “If you’re trusting a business process to this tech and it goes wrong one time in 30, 50, 200, that can be very expensive and costly to your brand, your finances and put you in breach of regulation.”

The Unlikely AI Solution: Accuracy, Explanation, Consistency

Unlikely AI’s approach centers on three core pillars that directly address the trust deficit: accuracy, explanation, and consistency. Unlike conventional large language models that often generate plausible but incorrect information, Unlikely’s platform guarantees correct outputs every time., according to additional coverage

The system provides thorough explanations for how it arrives at each conclusion and why the result complies with relevant regulations. Most importantly, it delivers identical answers to identical queries regardless of how many times the question is asked. “There’s nothing more trust damaging than seeing this AI system that you’ve built giving two different answers for the same data when you run it twice,” Tunstall-Pedoe emphasizes.

The High Stakes of AI Implementation

Current generative AI systems face significant challenges in enterprise environments. Tunstall-Pedoe cites an MIT report indicating that 95% of generative AI pilot projects at companies fail. The consequences of AI errors in business contexts extend beyond simple inconvenience to include legal ramifications, regulatory violations, and substantial financial costs.

Unlikely AI targets what Tunstall-Pedoe calls “high stakes industries where the cost of something going wrong is very high.” The company has already established partnerships with Lloyds Bank for customer support applications and insurance group SBS for claims processing automation. Another product automates accounting disclosures, a traditionally “massively tedious” task performed by junior accountants.

From Alexa to Unlikely: Building Trustworthy AI

Tunstall-Pedoe’s journey in AI spans decades, beginning with voice-activated search startup Evi, which Amazon acquired in 2012. His work ultimately contributed to the development of Alexa. After leaving Amazon, he knew he had “one or two big startups left in me,” leading to the creation of Unlikely AI.

The company‘s vision extends beyond immediate commercial applications. Tunstall-Pedoe aims to build “a very big, standalone business” that could potentially reach the trillion-dollar valuation that has eluded European tech companies. With nine U.S. companies valued above $1 trillion compared to Europe’s current highest valuation of €342 billion (ASML), the ambition represents a significant challenge.

The UK’s AI Ecosystem and Global Competition

Tunstall-Pedoe remains bullish on the UK as a technology hub, citing talent, world-class universities, and a friendly business environment. Regarding recent multi-billion pound investments from OpenAI, Google, and Microsoft into UK AI infrastructure, he sees potential benefits while acknowledging concerns about technological sovereignty., as detailed analysis

“There’s some reasons why that sovereignty piece can be important, and there are advantages to establishing British headquartered, very successful enterprises,” he notes. “But it’s not the case that just because an American company is investing that’s not a fantastically good thing.”

The Future of Enterprise AI Adoption

Unlikely AI’s pricing model reflects its confidence in delivering value: outcomes-based pricing where the company gets “paid per positive outcome.” While not currently fundraising, Tunstall-Pedoe anticipates needing additional capital next year to scale operations.

The company’s approach represents a fundamental shift in how enterprises might implement AI. Rather than focusing on flashy demonstrations, Unlikely AI prioritizes reliability and trustworthiness—the essential ingredients for meaningful business integration. As Tunstall-Pedoe concludes, “if you get the trust bit right you accelerate adoption.”

With major financial institutions and insurance companies already testing its technology, Unlikely AI demonstrates that solving the trust problem might be the key that unlocks AI’s true enterprise potential. The success of this approach could determine whether AI becomes a reliable business partner or remains an interesting but ultimately untrustworthy technology.