The AssumeAI blog: AI for consultancies and professional services

    What we arelearning, weekly.

    The frontier of AI, the practical development that applies to your firm, and the advances in the industry as a whole. A post every week.

    Over the next 24 months the pace of frontier AI development will increase, whilst the public will meet that frontier later, and on narrower terms, than it does today. The trend has arguably begun already, and I think it is, on balance, a good thing.Read the post ↘
    5 Oct 2026 · ai safety

    The frontier will move faster, and the public will see it later

    In late September, Amazon began quietly emailing some of the thirty thousand people it had laid off this year, asking whether they might like to discuss open roles in artificial intelligence. It produced one of the more honest statistics this year's AI layoffs have generated: the rate at which a company asks its own redundancies to come back.Read the post ↘
    30 Sept 2026 · ai adoption

    The most honest AI statistic this year is a rehiring email

    A military analyst asked a chatbot to summarise intelligence on a Chinese vessel, and the chatbot invented a nuclear weapons cargo that was not there. The report moved up the chain of command unchallenged until aircraft were already in the air.Read the post ↘
    23 Sept 2026 · ai agents

    The AI safety boundary most businesses trust is only an interface

    Dario Amodei spent last Saturday asking the AI industry to slow down. His essay, We Must Pace the Frontier, proposed that frontier developers give independent evaluators permanent, employee-level access to verify their safety commitments, and that the industry eventually accept limits on its most dangerous capabilities.Read the post ↘
    16 Sept 2026 · ai safety

    The case for trusting AI's safety claims got weaker this week

    OpenAI's newest model reasons in loops that leave no readable trace, at the same moment English law is building liability for AI harms on the assumption that a court can establish what a system did and why. The two trends are heading in opposite directions.Read the post ↘
    9 Sept 2026 · ai safety

    AI reasoning is going dark just as the law demands it

    Every part of this week's AI news describes the same shortage. Frontier labs are shipping agents into the most consequential parts of a business faster than anyone is building a reliable way to check what those agents actually did, and that gap, not the adoption curve, is now the more expensive one.Read the post ↘
    2 Sept 2026 · ai adoption

    AI agents are being trusted faster than they are being checked

    Everything in the media is polarised. That polarisation is not merely an aesthetic problem with journalism; it is creating a genuine and widening misunderstanding about what AI actually is, what it can do, and what businesses should reasonably expect from it.Read the post ↘
    15 Jun 2026 · ai consulting

    What the media and business press still get wrong about this AI transition

    AI is used more frequently in businesses than ever before, in recent studies 77% of consultancy firms in the UK have started to use AI in some way, yet on average 55% of the employees at these firms either do not adopt the technology or are using shadow AI.Read the post ↘
    9 Jun 2026 · adoption

    Why Adoption is the Most Important Thing

    On the 30th May, we were fortunate to be invited to the OXAI Frontier event at the Mathematical Institute, Oxford University, to speak about the future of work and how AI is affecting, and will continue to affect, companies of all sizes.Read the post ↘
    4 Jun 2026 · oxai

    Assume AI at the Oxford AI Frontier Forum

    When people ask which pattern in AI adoption concerns me most, they tend to expect something about the technology itself, but my answer sits elsewhere, and it is almost a double-edged sword. The pattern that worries me most is not the development of business software; it is the protection of it.Read the post ↘
    1 Jun 2026 · ai implementation

    The AI Risk Most SMEs Don't Know Is Just Around the Corner

    SpaceX can bring about another wave of advanced engineering and acceleration to the AI race, due to the complexity of the problems that must be solved in order for space missions to be possible and AI's ever-evolving ability to solve these problems.Read the post ↘
    25 May 2026 · engineering

    Space, AI, and the Next Wave of Inventions

    Those who manage a relationship with a client: your roles are safe. And therefore, how does one protect their job security in the world of AI? They develop relationships, maintain those relationships, and deliver relentless value.Read the post ↘
    18 May 2026 · ai transformation

    In the age of AI, relationships are the moat

    Book a strategy call