I FIRST spotted it as I trudged around Partick Cross, on the way to Kelvingrove, as a science-fiction-seized Glasgow University student in the early 80s.

“The Turing Institute” read the modest, modernist sign, attached to a standard business four-storey, its brown glass making it opaque to the world.

I knew who Alan Turing was – the Bletchley Park code-breaker during the Second World War, the first visionary of general artificial intelligence. So I wondered, with increasing feverishness, what this institute could conceivably be up to.

Alasdair Gray’s novel Lanark was just settling into the culture. So I fancied mysterious labs, ginning up militarised automatons, in service of the scientists of Unthank. An excitable child...

It turns out that the Turing Institute (well prior to the current Alan Turing Institute at the British Library) was an entrepreneurial venture by the late AI pioneer Donald Michie. The professor was a great chess-playing friend and associate of Turing during their war efforts.

Michie was trying to keep artificial intelligence alive, by selling industry and government its more practical solutions. The web (which the Turing Institute prototyped in 1988) tells me they developed code for Nasa’s Space Shuttle, for seed sorting at a Scottish agricultural agency – and much else besides.

READ MORE: Unionst MSPs embarrass themselves with outrage

All this comes to mind as a result of Edinburgh University’s celebration of their 60 years of artificial intelligence research, begun by Michie and others on their campus in 1963. It’s all promo for their School of Informatics (the concept under which Edinburgh gathers its digital research).

The university is brawling in the global battle for students, grants, talent and expertise. This is clear ammunition.

But what the anniversary justifies is yet another alert against current dolefulness about Scottish national prospects. Scotland, far from being a backwater in major trends for the future, has been involved at the core of their development. And can be again. It’s enough to start by noting the major figures in current AI that have passed through Edinburgh’s cyber-educating halls.

Two Googlers are most prominent. Geoffrey Hinton, father of the neural network approach, who has retired from an executive position recently to take up a Cassandra role warning about runaway AI.

There’s also Fernando Pereira, currently vice-president and engineering fellow at Google, who is a pioneer in natural language processing – that ability to respond to and understand human language that freaks us out about ChatGPT today.

The heydays (and the setbacks) of Michie’s career in Edinburgh and Glasgow provide a helpful historical anchor. We are often faced with a breathless “now” of radical technologies, with claims and ambitions coming out of nowhere to shock and thrill us. Yet it’s all technical progress towards some pretty ancient aspirations.

In a quite brilliant closing essay from his book’s overview of the field in 1986, Michie writes about the “first 2400 years of machine intelligence”. This isn’t an exercise in speculative fiction. Michie’s going backwards, as far as Socrates’ thoughts on the invention of writing.

In Plato’s Phaedrus, Socrates discusses an Egyptian god who has invented systems of calculation and writing. Another god, Ammon, makes what will sound like a familiar critique. “Those who acquire it will cease to exercise their memory and become forgetful; they will rely on writing to bring things to their remembrance by external signs, instead of on their own internal resources.”

And more on these infernal information devices (in his time, books; in our time, ChatGPT?).

“As for wisdom, your pupils will have the reputation for it without the reality: they will receive a quantity of information without proper instruction, and in consequence be thought very knowledgeable when they are for the most part quite ignorant … And because they are filled with the conceit of wisdom, instead of real wisdom, they will be a burden to society”.

As Michie writes, with amazing prescience: “The objection here is concerned with the simulation of knowledge by the possessor of a rapid-access source. Precisely because the distinction can be blurred between possessing a sufficiently fast knowledge-source and actually possessing the knowledge, we see Ammon’s point as not only subtle but also topical.”

As topical as any educator you might meet, wringing their hands about student essays written by chatbots.

Michie’s erudite essay also quotes Jonathan Swift’s Gullivers’ Travels, in which a professor recommends the use of a “generate-and-test treatise-writing machine”.

From Swift: “Everyone knew how laborious the usual method is of attaining to arts and sciences; whereas by his contrivance the most ignorant person at a reasonable charge, and with a little bodily labour, may write books in philosophy, poetry, politics, law, mathematics and theology, without the least assistance from genius or study.”

Swift’s satire is now our reality.

READ MORE: Mhairi Black tears apart Tories' Rwanda policy on Question Time

SO we’ve been dreaming of (as well as slightly fearing) machine intelligence becoming our mental assistant for a long time.

As Michie says at the end of this piece, Alan Turing’s 1947 notion of a “Universal Machine” (our idea of a powerful AI) was remarkable “not for its programmability, but for its self-programmability”.

As Turing wrote: “It would be like a pupil who had learnt much from his master, but had added much more by his own work. When this happens I feel that one is obliged to regard the machine as showing intelligence … As soon as one can provide a reasonably large memory capacity it should be possible to begin to experiment on these lines.”

This is precisely what has happened in the current upsurge in AI. The sector has experienced a massive expansion in memory power, or “compute” as the techies put it. Vast amounts of information drive the perception of sophisticated patterns.

But these calculations are so complex, so unimaginably simultaneous, that our current AIs are like black boxes or even mystic portals. We see them “auto-complete” our answers in Chat-GPT4, but no-one has any idea how it actually happens.

Turing anticipated this in 1947, quoted by Michie: “The instructions have been altered out of recognition … but nevertheless still are such that one would have to admit that the machine was still doing very worthwhile calculations.”

Michie’s work also encompasses that part of automation which should give us humans (at least temporarily) some scrap of comfort. And that’s the terrible quality – and stuttering progress – of human-like robotics.

There’s an early 70s TV science debate on YouTube – barely distinguishable from a Fast Show parody – where Michie is on the lead panel, pushing back against the UK withdrawal of funding from AI research.

By all historical accounts, it was this transmission that decisively triggered what has been termed the “AI Winter”. This is a transatlantic era (decades-long) where artificial intelligence was starved of resources.

Michie possesses the best of Anglo-Edinburgh eloquence: born in Rangoon, father a banker for the Empire, languid RP voice. But he’s never going to win the day with the video he shows the programme, as justification for his AI work.

It’s a giant robot pincer, with dials for eyes and a grille for a mouth, called Freddy. The demo involves scattering pieces of a toy car randomly across a cloth and leaving Freddy to assemble them correctly.

What a two-year-old could do in minutes before cheese and chips for lunch, is for Freddy the very labour of Sisyphus. Slowly, like the dullest creature on earth, and over unspecified hours and days, the car is put together.

Human-substituting, it isn’t.

Of course, industrial robotics works with humans, in centaur-like relations, across the world now – and that’s been Freddy’s path-breaking contribution. But it’s still an entertainment to encounter even the best of humanoid robotics, tentatively shuffling about their environment like wizened elders, as autonomous as the mighty battery packs and cables they’re connected to.

So we’ve much to be thankful for Michie, and the University of Edinburgh (wishing the latter well in its engagements with the AI talent wars). Certainly for anticipating and preparing for our current moment of vertigo. But also for reminding us of the limits of the techno-future.

Of course, I wish I’d buzzed that mysterious door on Dumbarton Road…