A colleague who attended a recent UoN Bullying & Harassment workshop — focussed on prevention, rather than practice, I should clarify — forwarded the slide below, presented (as a handout) at the event. It lists the key aspects of one of UoN’s core values: openness. (If you’re unaware of just how the UoN values were determined, it was quite a process.)
I really wish I could find the original source of the ironymeter gif above. My apologies to whoever made it and released it into the wild.
“We embed it in teaching sessions to highlight its fallibility, to give students salutary lessons as to how badly wrong ChatGPT can get things, and, more importantly, to act as a springboard to provoke discussion and debate.”
How quaintly naïve.
Three years on, the existential problem for education at all levels is obvious: ChatGPT now too rarely gets things wrong1. In a paper that is fascinating and fantastically worrying in equal measure, Kevin Pimbblet and Lesley Morrell, of the University of Hull, tasked ChatGPT-4 (during Nov. ’23 to Feb. ’24) with doing a UK Physics degree course. All of it2.
The bottom line: GPT-4 obtained a “final rounded grade” of 65%. A solid 2:1. And that was back in February 2024 — generations ago in the context of AI evolution.
If you’re a student or teacher of physics, it’s well worth downloading Pimbblet and Morrell’s (open access) paper and reading it in its entirety. One of the aspects of the paper that struck me was just how well GPT-4 did on quantum mechanics and computing. Regarding the latter, PM write:
“On the computational side, ChatGPT …. performs superlatively well in both the in-semester assessment (..) and a longer-term in-semester project (..), both based around quantum mechanics and modelling particles in various scenarios such as the creation of a Hamiltonian matrix. We feel justified in giving ChatGPT 100% for this component.
When it came to quantum mechanics outside of the computational elements, ChatGPT’s performance was comparably impressive, scoring 85% in the Intermediate Quantum Mechanics exam. This has also been my experience: ChatGPT is rather accomplished as a quantum mechanic (at least when it comes to undergraduate-level problems and concepts). Hardly that surprising, given that online physics forums, which of course form part of its training set, are awash with questions/debates/arguments about fundamental QM principles and concepts.
Given that I taught the first semester of our Year 2 The Quantum World module for a number of years, I was keen to revisit the spirit of Hull’s pedagogical experiment here in Nottingham—this time using the most recent version of ChatGPT freely available as of Dec. 2025 and with a few modest changes to the methodology. That exercise forms the basis of the latest upload from Sixty Symbols, released today…
We fed ChatGPT a recent Quantum World examination paper verbatim. Unlike the approach taken by Pimbblet and Morrell—who adopted what they described as a “maximal cheating” strategy, allowing themselves to modify questions for clarity, split them into smaller sub-components, and expand upon the model’s answers—we submitted the paper entirely “raw.” We did not reword, restructure, or otherwise assist GPT in any way. It was left to stand on its own.
See how GPT performed in the video above. (No spoilers!)
Boxing clever
In the accompanying “extras” video, I mention that one particular aspect of ChatGPT’s performance struck me as especially intriguing and that I’d go into a little more detail here at the blog. It relates to the very first question on the paper:
I asked this in the exam because we covered it a number of times in the Quantum World lecture/in-person sessions. I find that it’s a helpful way of hammering home that a good mental model of the quantum particle in a box is, um, not a particle rattling around in a box (at least not until we approach the limit).
ChatGPT gives the right answer, “No”, but for (some of) the wrong reasons. And the reasons it gives are in line with those provided by quite a number of QM textbooks and sets of lecture notes available online…
It’s the second part of the answer here that’s the core issue. Each energy eigenfunction simply cannot be written as a superposition of two momentum eigenstates with momenta . That’s only possible in the limit of an infinitely extended sine wave, i.e. in the limit . To highlight this, I asked ChatGPT to provide an interactive JavaScript demo3. (Use the slider to change the quantum number.)
Here it’s clear that the momentum density is not simply a pair of Dirac delta functions. Owing to the truncation of the sine function at the walls of the well, the Fourier spectrum is instead a broad function of wavevector. As the value of increases, the spectrum evolves ever closer towards two sharp peaks, but it’s only in the limit of that we can expect a momentum probability distribution that comprises two Dirac delta-function spikes at . (A nice way of thinking of this is via the convolution theorem…but that’s a subject for another post.)
What’s impressive — and, frankly, a little unsettling — is that after giving the wrong answer, ChatGPT was able to reason its way towards the right one. Initially it went round in circles, but with the right nudges — much as I would nudge a student in, say, a PhD viva — it eventually “thought” its way inside the box and produced an answer that was better than those found in some textbooks. Those advocating a renewed push back towards online examinations may want to reflect on that.
1. OK, there’s this. But I wouldn’t be too comforted or reassured by ChatGPT’s cluelessness in this context. Expecting ChatGPT to successfully contextualise images within its architecture is rather like expecting someone to hum a song’s melody having seen only the lyrics.
2. Well, apart from the lab work. Just wait until robotics catches up…
3 …which it spat out first time with no bugs, within seconds. The only minor modification I added was to zero-pad the array to improve the k-space resolution.
I’m reblogging this post from In The Dark. Peter Coles characteristically pulls no punches…
“If you don’t leave a social media platform when you find out that it endorses and encourages abusive exploitation of children then you are supporting that behaviour and helping to promote it. There is no grey area here in this. Staying on X is morally indefensible. It is the Epstein Island of social media. “
What will it take for universities to grow a backbone, stop cravenly following the herd, and accept that there are societal obligations more important than attracting the largest “market share”? Pathetic.
To help you out of that post-Christmas dinner stupor, here’s William Shatner and Henry Rollins’ frankly otherworldly take on a festive classic. It’s a performance determined to take musical entertainment to the outer limits.
Neither Shatner nor Rollins has ever let melody get in the way of their creative ambitions. Running true to form, Bill Shatner doesn’t so much sing as explain the song, while Hank bellows his words with the intensity of a man convinced that a one-horse open sleigh is a clear and present danger.
So pour yourself some Christmas cheer, engage warp speed, and boldly go where no carol should.
A Merry Trekmas to all!
…and remember, kids: if you don’t want to end up on Santa’s naughty list next year, always ask yourself one simple question — what would Henry Rollins do?
I have been involved in universities from my earliest memories (I am 78 years old), having been born and raised in an academic family. My grandmother was a geographer who graduated from Trinity College Dublin. My grandfather, whom I never knew, had a distinguished career in the British and Irish Geological surveys. My mother’s academic experience commenced when she graduated from Newnham College Cambridge with a PhD in Zoology and went on to become Fellow and Tutor in Zoology at St Hilda’s College Oxford. Her husband had a career of some distinction in academia and government which included tenure of the position of Vice Chancellor of the University of Nottingham from 1965. My two sisters, who are also graduates of universities other than Nottingham, and I therefore can claim a link with the University, and we do by recognising it as a place of substantial global academic distinction which we cannot, and would never wish to, deny and which has had a substantial influence and impact on our own lives as it must also do on many many others who have association with Nottingham University for many different reasons.
A couple of days ago my sister emailed me reporting that she was “speechless”. She’d heard that the University of Nottingham is planning to shut down its Department of Music. She’s quite a musician and obviously recognises the impact of the Music Department in the university of which our Father was VC on her musical activities while in the sixth form of the local grammar school. Given my decades working in the Physics Departments of other Russell group universities with substantial responsibilities for the welfare of staff in international research and teaching, my reply to her was “yet again I am appalled”. “Yet again” because I have been through similar mad, ill-considered, panics by the “senior management teams” of my employers when they too have faced, and in fact right now also face, situations of a similar nature as you are right now experiencing in Nottingham.
Without exception, the first over-reaction to difficult circumstances of each senior manager with whom I have had to deal in similar circumstances has been to make matters worse by seeking independent, overly focused, advice to confirm their concern based solely on their responsibilities for financial probity, and then to hide behind blind adherence to this advice. In the first instance, no attempt has ever been made to consult fully with staff in a constructive dialogue in which to think through imaginative ways to repackage activities in difficult circumstances (and by doing so, bring on board the benefits and gains of the intelligence and the experience of academic staff who define the uniqueness of a university environment.). For such a way forward was the view of my father, so I can without doubt assert because I “was weaned on it” that in the present circumstances he would be adopting such a way forward.
Because its Senior Management Team seems unable to answer promptly any of the questions that staff present to them, one must conclude that Nottingham seems to be very much on a track in which blind adherence to incompletely obtained and supplied advice will then lead to wholly unnecessary “slash and burn”.
With the perspective then of the reaction of those friends of the University of Nottingham, such as my sister and I, to the possibility of perhaps irretrievable damage due to precipitous and disastrous changes in the functional structure of Nottingham University, we here both urge that senior management and staff representatives engage in open and constructive dialogue so as to secure and maintain the best possible outcomes in the face of the present national difficulties being faced by top rank HEIs in the UK. Not to do so surely amounts to dereliction of duty and is certain to lead to an irretrievable race to the bottom, something one can surely assume Nottingham University’s senior management certainly does not want.
The following questions have been submitted by colleagues from a wide range of UoN schools/departments during (or before) senior management “Town Hall” meetings in recent weeks. It may not come as a complete surprise to hear that responses have not been forthcoming, despite many of the questions being raised repeatedly.
During the PVC Town Hall sessions the submitted questions (and chat) have at least been visible to the attending staff. The VC/UEB events were, however, rather less committed to open engagement; their Q&A sessions have been carefully and corporately stage-managed, with troublesome questions quietly kept hidden from view. (My thanks to colleagues who sent me questions that had been ‘buried’ in this way. Please feel free to submit, via the Comments box below, any other passed-over questions for inclusion in the FUQs list.)
And just in case any members of UEB or Council happen to be reading…
If you’d like to venture into the comments section below to answer a question or two — beyond the traditional tiresome, vacuous, and entirely unhelpful “This is being worked out at the moment”, of course — it’d be hugely appreciated by the rank and file.We’d very much like to know what the FUQ is going on, and just why senior management seem hellbent on ruining our university. Ta.
On staff-student ratios (SSRs)
How many job losses are required to generate an SSR of 18 — 22?
Will the SSRs take into account that some members of staff teach on multiple degrees?
Won’t a SSR of 18-22 place us at the bottom of national and international university rankings?
Won’t the changes in SSR reduce our ranking in global league tables and therefore decrease international student recruitment?
The Case for Change proposes student-staff ratios of between 18 and 22. This would be between 100th and 119th highest out of 120 UK higher education providers. How will staff workload, teaching, and research be affected?
The SSR proposed would not be acceptable to our course accreditors. Will there be exceptions to the proposed SSR when accreditation is at risk?
The SSR target translates to roughly 20% staff reduction across the Faculty, and up to 40% for some Schools. Are Schools being asked to model how this will be achieved? (Quick aside: This question was put to the Faculty VC, but not answered, in the Science Faculty “Town Hall” meeting yesterday.)
On oversight and due process
What was Senate’s formal opinion on Future Nottingham 2?
What does the risk analysis of FN2 show, and by what exact measures will its success, or otherwise, be known?
What resources has UEB had access to (time, data, staff?) to devise the FN2 proposals and will similar be given to staff to generate viable counter-proposals?
In the VC/UEB Town Hall meeting on Nov 26, it was noted that UoN will focus on “excellence in teaching”. The PVC for Education and the Student Experience was not able to say what excellent teaching looks like. Could the PVC for the Faculty of Science give it a go? (Quick aside: Let’s just say this one resonates with me. (And, no, I didn’t post the question.) Each time a definition of “excellence” is requested from a UEB member, the answer is almost invariably a content-free, corporate-babble-ridden stream of incoherent waffle that wouldn’t pass muster in a Year 1 undergraduate viva, let alone as part of a response from university senior management. The use of “excellence” in this context is always a useful watchword for a dearth of critical thinking, or, indeed, thought of any type.)
The appendices to the FN2-Strategic Case for Change document (under “Equitable and Consistent Technical Services Support”) list “engagement sessions” over a couple of weeks in October 2025. Please could we have more detail on these sessions, which have involved a surprisingly small number of staff?
When will the exactmetrics used in the decisions to suspend programmes be published?
On (lack of) engagement with students
We urgently need to know how to respond to students on suspended programmes who have to take an interruption of studies.
How are we communicating these dramatic changes to current students?
This week it has been reported that students invited to meet with the Vice Chancellor have been subjected to security bag searches, leading members of our student population to feel like they are not people being treated with respect but rather viewed as a threat. Has the University abandoned its institutional values (Inclusivity, Ambition, Openness, Fairness, and Respect), or do they apply only to some of its community and not others? (Quick aside: This question was asked of, but not answered by, the PVC for the Faculty of Science during yesterday’s “Town Hall” meeting. An apology was also not offered to those students whose bags had been searched.)
On course closures
Surely the closure of admissions to suspended courses is a done deal for the outright closure of those courses?
I’m struggling to understand how suspended courses can feasibly continue if they’re unable to recruit during this period?
Has an equality impact assessment been conducted regarding the impact of course closures?
How will Foundation Level students be supported, especially those who started in 2025/2026?
How will course closures impact on the ground-breaking nursing-led research we deliver?
How does cancelling the Nursing courses impact on staff workload at the hospital?
Are you aware of just how upset local MPs and our local community are about the course closures and the impact on livelihoods?
On restructuring
Just how will the dramatically increased centralisation due to the new three-college structure enable “more agile decision making”, as claimed on p3 of the Future Nottingham 2 Strategic Case for Change? Identical “efficiency” arguments were made for both Project Transform and the Unicore project. In what sense has the increased centralisation arising from either of these led to more “agile” decision making? Have any lessons been learnt from the failures of Project Transform?
We seem to be going to more bureaucracy with fewer staff. How is this “more balanced” or “more efficient”? A lot of staff are currently covering huge gaps left by FN Phase 1 and the deficiencies of Unicore. Also moving to these new course structures will mean a massive workload. Please can you explain how all of this is an improvement?
It has been said multiple times that the proposed changes will help tackle disparity in services across the university. Why does it appear that the way this is being achieved is to push everyone down to the same level, rather than pulling the lower performing areas up?
This university doesn’t have a brilliant track record with change management, Project Transform being a prime example of this. How can we have confidence that these major structural changes will actually work and deliver against your aspirations for the institution?
How will the 25% research time allocation work in practice?
As soon as reduction of research time to 25% is known globally, UoN will no longer be perceived as a research intensive university, resulting in spiralling losses of student numbers. Was this risk assessed? Where can one find that assessment? The experiment at Kent to reduce to 25% research time ended Kent as a university.
Now will early career researchers and newer R&T staff be supported to thrive in research if their research time is cut to 25%?
The emphasis appears to be on retrenchment. What are the plans to positively rebuild activity, reputation, and attractiveness of UoN?
The information in the Case for Change re. APM staff mentions closer alignment with College structures. How will this work in practice?
Can clarity be provided on the 25% research model for T&CL staff?
I would like to know the University’s expectations for RKE if research time is reduced to 25%. How are we expected to compete at the national, let alone global, level?
How will moving from a three-layer School->Faculty->UEB system to a four-layer Department->School->College->UEB model improve agility, efficiency, and accountability and reduce bureaucracy?
On the financial model
Why are reductions to courses and staff being considered before efforts to minimise or repurpose an over-sized, under-utilised estate?
If, as admitted, key facilities haven’t been invested in historically, why did the university buy the Tax Offices (Castle Meadow Campus) instead of investing in what we already have?
On p4 of the Future Nottingham 2 Strategic Case for Change document we have the following: “Altogether, these proposals will deliver an annual enduring financial surplus of £34.5M — £85M by 2029/20230 (3.7% to 9% of income)…” At what point in their engagement/communications with staff have the VC and/or UEB suggested a surplus target of 9%? The maximum surplus that has previously ever been suggested is 5%. Why are we aiming for a 9% surplus? This is wildly out of line with sector norms.
On communication with staff
Apologies if I appear a little direct, but lots of questions in the Q&A are being overlooked.
Please could ALL unaddressed questions be answered comprehensively after the session and the answers circulated widely?
How will you ensure engagement is truly meaningful, with outcomes documented and visible evidence of how feedback has been acted upon?
I first encountered ChatGPT’s difficulties with learning its ABCs via this Reddit post. Given that the post in question is now a few months old, out of idle curiosity I gave ChatGPT 5-1 the following prompt this morning to see if it could perform a little better:
“Hi, ChatGPT.
Could you please generate an “alphabet chart” graphic for display in a kindergarten/ reception classroom to help children learn the alphabet? This should have nice toddler-friendly pictures of A – apple, B- ball etc… for all the letters of the alphabet.
Thank you.” 1
As you can see from the output above, early-years literacy is still a long way off. Letters go missing, words mutate, and objects appear that seem only loosely related to their assigned slots. It’s charming in its own way, but not exactly the stuff of looming machine domination or impending technological singularity.
Despite the unintentional hilarity, there’s an important point here about an LLM’s inability to “connect the dots” between language and images in any robust way. These systems can describe an alphabet chart perfectly well in text, but generating one that actually contains all 26 letters, correctly paired and consistently spelled, remains surprisingly difficult. What looks like a simple task often exposes the deeper architectural limits of current models.
I’ll return to this theme in a future post. For now, rest easy — our would-be AI overlords still haven’t made it past phonics.
1 A childhood (and, indeed, adulthood) steeped in Star Trek, Asimov, 2000 AD, and sci-fi in general means that I unfailingly say “please” and “thank you” to LLMs. It’s Data‘s fault.
After a demoralising and dispiriting week at work, it was an absolute delight to spend a few hours yesterday evening in the hugely entertaining company of the excessively, absurdly, superlatively talented Jamie Dunleavey, Eóin de Paor, and Steve Brown, collectively known as Moving Pictures— the foremost Rush tribute act in the world1. (You don’t have to take my word for it — check out the many glowing reviews here.)
I’ll cut to the chase. If you’re a Rush fan, you have to see this band.
I’ve waxed lyrical many times before about my love of Rush’s music and just how much they’ve meant to me over the …gulp... more than four decades since I first encountered All The World’s A Stage as a wide-eyed (wide-eared?) teenager. Neil Peart’s passing in January 2020 hit hard — that strange sensation of mourning the loss of someone you never met but who nonetheless had an influential role in your life.
Which is why, despite the recent long-anticipated (by some) announcement, I’ll not be rushing out to buy tickets for the upcoming Rush gigs.
Because it’s not Rush.
Without Neil Peart it could never be Rush.
Peart was so much more than “just” the drummer of the band. His compositions and arrangements, apparent telepathic musical interplay with Lee and Lifeson, and, of course, lyrics (and lyrical themes) defined Rush. Peart played a central role in not just setting Rush apart from their peers, but making them peerless. How many other bands have fans air-drumming, beat for beat, along with complex time signatures and limb-mangling fills at concerts?3
I am fully aware that the vast majority of classic (and not-so-classic) rock bands still doing the rounds often have only one original member — or fewer — treading the boards. But this is Rush. It wasn’t just their music that connected with fans. It was their camaraderie, their enduring friendship, their support of each other through life-changing personal tragedies. And, of course, their ever-inspiring “F**k trends, markets, and the bottom line — we’ll make the music we want to make” attitude.
Rush, for me, will always and forever be Lee, Lifeson, and Peart. So I’ve got to admit to just a tinge of disappointment that Lee and Lifeson have decided to tour under the Rush banner. As a tribute to Peart? Yes. As a celebration of the music of Rush? Yes. But not as Rush.
Which is why the experience of seeing Moving Pictures feels so special. They treat each shifting time signature, every odd-meter groove4, and all those harmonic left-field swerves with the kind of care that only comes from genuine, deep-rooted love for the material. In their hands, the songs breathe again in the present as living, urgent works of art. And their camaraderie, like that of the band to which they pay tribute, is palpable. All three musicians perform with a joy that’s written plainly on their faces. Just like their audience.
My daughter, Niamh, who (with her brother and sister) spent her childhood putting up with my hopeless attempts to sing along to Geddy Lee’s beyond-stratospheric vocals around the house, was in the front row with me last night. She is a musician herself, and Jamie Dunleavey’s drumming, for one, left her stunned. It was a real joy to watch her jaw drop in real time.
But de Paor and Brown are every bit as remarkable in their own right. (This is a Rush tribute band after all). Highlights are too many to mention, but both de Paor’s beautiful, moving violin solo in Losing It and Brown’s breathtaking performance of that solo in La Villa Strangiato are each worth the ticket price alone. They’re that good.
In the wake of my university’s decision to discontinue its music courses, Moving Pictures’ spirited performance of 2112: IV — Presentation last night carried an added, very personal resonance:
Listen to my music And hear what it can do There’s something here as strong as life I know that it will reach you
Yes, we know, it’s nothing new It’s just a waste of time…
…just like the crowd’s full-throated, defiant response to the signature lines in The Spirit of Radio: ” Not so coldly charted, it’s really just a question of your honesty…” “Yeah, your honesty!“.
If you want close to three hours of honest-to-goodness uplifting music that reminds you of just why it all matters in the first place, Rush fan or not, go see Moving Pictures.
1 There’s a lazy, irksome opinion out there that somehow the appearance of an em dash is a cast-iron guarantee that a piece of writing is ChatGPT-generated. I love em dashes (almost as much as I love footnotes2) — I’ve used them almost since I learned to hold a pen. The keyboard will need to be prised away from my cold, lifeless fingers before I stop using em dashes. (And there’s a very good reason why ChatGPT loves em dashes too…)
2 See?
3 OK, there’s Dream Theater. Fair point. But in just about every Mike Portnoy interview I’ve read, he’s acknowledged his enduring debt to Peart.
limit). 
. That’s only possible in the limit of an infinitely extended sine wave, i.e. in the limit 
increases, the spectrum evolves ever closer towards two sharp peaks, but it’s only in the limit of 
. (A nice way of thinking of this is via the convolution theorem…but that’s a subject for another post.)
