Loading organizations...
§ Private Profile · London, United Kingdom
Unlikely AI is a technology company.
Unlikely AI builds artificial intelligence solutions for highly reliable, auditable decision-making. Its core product ensures consistent, verifiable outputs, with identical inputs always yielding identical results. The company specializes in inherently controllable, hallucination-free AI systems, directly addressing the critical demand for precision and trustworthiness in enterprise applications.
William Tunstall-Pedoe, creator of Evi and co-creator of Alexa, founded Unlikely AI in 2019. He recognized limitations in large language models and the critical need for trustworthy, safe, and controllable general AI. His ambition was to build deep technology delivering effective, reliable AI for real-world applications.
Unlikely AI targets sectors demanding high-stakes decision reliability, notably within the insurance industry for precise claims processing. Its solutions empower businesses to leverage AI they comprehend and trust, fostering operational confidence. The company envisions establishing a new standard for AI-driven decision-making, enabling organizations to act with greater assurance and transparency.
Unlikely AI has raised $20.0M across 1 funding round.
Unlikely AI has raised $20.0M in total across 1 funding round.
Unlikely AI has raised $20.0M in total across 1 funding round.
Unlikely AI's investors include Octopus Ventures, Amelia Armour, Amadeus Capital Partners, Cambridge Innovation Capital, Kayne Anderson Capital Advisors, L Catterton Growth, Ben Huh, Specialist VC, Balaji Srinivasan, Dylan Field, Barak Berkowitz, Christopher North.
Unlikely AI has raised $20.0M across 1 funding round. Most recently, it raised $20.0M Seed in September 2022.
| Date | Round | Lead Investors | Other Investors | Status |
|---|---|---|---|---|
| Sep 1, 2022 | $20M Seed | Octopus Ventures, Amelia Armour | Amadeus Capital Partners, Cambridge Innovation Capital, Kayne Anderson Capital Advisors, L Catterton, BEN HUH, Specialist VC, Balaji Srinivasan, Dylan Field, Barak Berkowitz, Christopher North, Patrick Pichette, Metaplanet | Announced |
Unlikely AI is a London-based deep tech startup developing a neurosymbolic AI platform that combines machine learning with symbolic computing to deliver trustworthy, explainable AI systems free from hallucinations, bias, and black-box decisions.[1][2][3] The company builds products like its Universal Language (UL) precision engine, which provides definitive yes-or-no answers with 99%+ accuracy, complete audit trails, and seamless integration for enterprises—solving reliability issues that plague traditional LLMs in high-stakes tasks.[1][3][5] It primarily serves regulated industries such as banking, insurance, finance, healthcare, and accounting, where trust, transparency, and compliance are non-negotiable, enabling automation without manual verification.[1][2][3] Growth momentum includes a $20M oversubscribed seed round led by Amadeus Capital Partners and Octopus Ventures, customers like Lloyds Banking Group and SBS Insurance, 16 patents in AI, natural language processing, and data management, and a pivot from R&D to market-ready products over the past year.[2][3][4]
Unlikely AI was founded around 2018-2019 by William Tunstall-Pedoe, a serial entrepreneur with a proven track record in AI and natural language understanding.[1][2][4] Tunstall-Pedoe previously built Evi Technologies (formerly True Knowledge), a natural language startup acquired by Amazon in 2013, whose technology became foundational to Amazon Alexa.[1] The idea for Unlikely AI emerged from his recognition of persistent flaws in modern AI—like hallucinations and lack of trustworthiness—despite advances in LLMs, prompting him to pioneer neurosymbolic methods years into deep R&D.[1][3][4] Early traction built on this expertise, with the company evolving from foundational research into a commercial platform, securing venture funding and high-profile clients amid rising enterprise demand for reliable AI.[3][4]
Unlikely AI stands out in the AI landscape through its proprietary neurosymbolic technology and enterprise focus:
Unlikely AI rides the neurosymbolic AI trend, addressing the "trillion-dollar trust barrier" holding back enterprise AI adoption amid massive infrastructure buildouts for LLMs.[3] Timing is ideal post-EU AI Act, which mandates explainability in high-risk systems (fines up to 4% of turnover), positioning neurosymbolic tech as essential for regulated markets where black-box AI fails.[3] Favorable forces include surging demand from finance/insurance for automation without errors, competition from IBM/EY/AWS validating the approach, and Unlikely AI's edge in proprietary precision for tasks like claims or credit decisions.[3] It influences the ecosystem by accelerating responsible AI at scale, enabling broader enterprise uptake and setting benchmarks for hallucination-free generative AI.[1][3][4]
Unlikely AI is poised to capture a slice of the exploding trustworthy AI market by expanding from insurance/finance pilots to full enterprise rollout, leveraging its $20M war chest for product scaling and talent hires in London.[4][5] Key trends like stricter global AI regulations, hybrid neurosymbolic adoption by Big Tech, and demand for auditable LLMs will propel growth, potentially mirroring Tunstall-Pedoe's Alexa success. Its influence could evolve into the foundational layer for compliant AI infrastructure, unlocking trillions in stalled value as enterprises prioritize certainty over raw power—proving that Unlikely AI makes the improbable reliable.[3][4]