I was amused to note this in the FT this morning, given my observation on the accuracy of ChatGPT yesterday:
Big Four accountancy firms are racing to create a new type of audit that verifies the effectiveness of artificial intelligence tools as they seek to profit from clients' demand for proof that their AI systems work and are safe.
Deloitte, EY and PwC told the Financial Times that they were preparing to launch AI assurance services as they hope to use reputations gained in financial audits to win work assessing whether AI systems, such as those in self-driving cars and cancer-detecting programmes, perform as intended.
Perhaps they'd like to start with working out why ChatGPT is happy just making stuff up? I should think these firms could make a fortune out of that. The cash registers must be rolling, almost out of control in their HQs at the prospect of this market, unless, of course, the makers of AI tools realise that this is a form of assurance that they cannot provide because it would be too embarrassing to do so.
In case it helps, this is the question they need to ask the AI search engine:
Did you just bullshit me?
I think that should do.
Thanks for reading this post.
You can share this post on social media of your choice by clicking these icons:
There are links to this blog's glossary in the above post that explain technical terms used in it. Follow them for more explanations.
You can subscribe to this blog's daily email here.
And if you would like to support this blog you can, here:
AI is a tool we can’t ignore, but it’s also deeply disruptive and helps burn the planet.
I for one won’t be using it until it’s properly taxed.
But who will audit PWC, E&Y, & Deloittes? Why would I trust THEM?
It should be noted that “AI” is an amorphous umbrella term for a lot of technology.
ChatGTP and it’s friends are generative AI. They use statistics and huge amounts of data to generate content.
There are other types of AI. For example image classifiers which can be trained to detect disease and are faster and better than humans. The same tech can be applied to finance to detect fraud and works reasonably well. They are generally very useful tools that have applications that have not been thought of yet.
However it all gets lumped into “AI” so as the buzz around generative AI dies, other useful tech is being thrown on the same bonfire
Agreed
I obliquely made that point here in response to a comment yesterday
When I was at agricultural collage in the 60’s a commonly used expression was – “BULLSHIT BAFFLES BRAINS.”
I haven’t heard it used recently, so thought it was a good idea to resurrect it
It seems to me to be a common technique these days, used by politicians, internet, social media, AI, you name it.
The companion phrase, IIRC, was “If you can’t convince them, confuse them.” Seems like generative so-called-AI is very helpful with them both.
For further comic relief, I offer the following anectode from yesterday’s *Guardian*:
‘The novelist Ewan Morrison was alarmed, though amused, to discover he had written a book called Nine Inches Pleases a Lady. Intrigued by the limits of generative artificial intelligence (AI), he had asked ChatGPT to give him the names of the 12 novels he had written. “I’ve only written nine,” he says. “Always eager to please, it decided to invent three.” The “nine inches” from the fake title it hallucinated was stolen from a filthy Robert Burns poem. “I just distrust these systems when it comes to truth,” says Morrison. He is yet to write Nine Inches – “or its sequel, Eighteen Inches”, he laughs.’
The whole article is well worth your time: https://www.theguardian.com/technology/2025/jun/03/creatives-academics-rejecting-ai-at-home-work
🙂
Gosh!
I am shocked …. And saddened. I had thought the limit was 12″ — but never to use it, as a rule.
🙂
Haha brilliant question, and wide ranging applications.
Will you patent it? 🙂
Thanks….
Just read a news piece about the bankruptcy of an AI company where the AI was apparently 700 people on computers in India. I believe it had originally secured over a billion in funding