I had another exchange with ChatGPT yesterday morning. Here’s the gist of it (I’ve excluded all ChatGPT’s detailed reasoning):

Me:

At last month’s Labour Party conference Ed Miliband, the Energy Secretary, is reported to have said: ‘Today, bills remain high because wholesale gas prices are still 75pc higher than before Russia invaded Ukraine’. Is that correct? 

ChatGPT:

 Yes, the statement is broadly correct in the sense that wholesale gas prices for UK households are still significantly higher than before the Russia-Ukraine war, and the figure of +75% is used in official commentary.

 But the figure is simplified, and one should treat it with caution:

• The baseline (“before Russia invaded Ukraine”) is somewhat loosely defined.
• Some independent data suggests the multiple is larger (or that it varies by contract/futures).
• The linkage to bills is not entirely one-to-one, since other costs also feed into bills.

Me:

It seems to be agreed that the UK’s energy prices are the highest in the developed world (if not beyond that). If that’s correct and in view of the above, presumably the main reason for that is the level of wholesale gas prices. Is that correct?

ChatGPT:

 Yes, you are broadly correct: the high level of wholesale gas (and gas-influenced electricity) prices is the key driver of the UK’s relatively high energy bills.

 But you should add the nuance that:

• other cost components matter (network, levies, taxes, supply margins)
• the UK’s particular system makes it more exposed to these wholesale gas swings (via generation mix, marginal pricing)
• for pure gas consumption, UK household prices are not always the highest globally; the electricity side is more clearly exceptional.

Me:

I find it hard to reconcile the above analysis with these Government data: https://www.gov.uk/government/statistical-data-sets/international-domestic-energy-prices. As expected, Table 5.3.1. shows that the UK’s industrial electricity price is well above that of our international competition. Yet Table 5.7.1 shows that the UK gas price is about average. 

ChatGPT:

Although UK industrial gas prices are broadly in line with international peers, the UK’s relatively high industrial electricity prices reflect the fact that gas‐fired power stations set the wholesale electricity price in the UK far more often than in many European countries. Thus elevated global gas costs feed through into electricity, which many industries rely on — pushing UK electricity costs above the average even if direct gas fuel costs are not outlier.

Me:

If, as you say, the UK’s electricity prices are so high largely because the UK electricity market is heavily exposed to gas‐fired generation, Miliband is right to say that we must end our reliance on gas and focus instead on renewable energy. Is that correct? 

ChatGPT:

Yes — Miliband is right in principle that reducing reliance on gas is essential to stabilise and eventually lower UK electricity prices. But that transition depends on ensuring enough low-carbon, flexible generation (and storage) to keep the lights on when renewables aren’t producing.

Me:

I don’t think your caveat goes far enough. Some knowledgeable commentators point out that it’s precisely because we have more renewables than our competitors (in percentage not absolute terms) that our energy prices are so high. They cite the system costs of renewables (e.g. subsidies, carbon taxes, grid balancing, grid expansion, ‘constraint’ payments (compensation for having to switch off) and the cost of back-up to cope with intermittency) plus the costs of updating the grid, of investment in ‘green’ hydrogen, of developing Carbon Capture and Storage for fitting to gas-fired power plants used as back-up. They conclude that all this means that adding yet more renewables, far from solving the problem of high energy costs, can only make things worse.

Are they wrong?

ChatGPT:

The critics are right that the current mix of renewables, subsidies, and system costs pushes UK prices up. But that’s mainly because the system is halfway through a costly transition — not because renewables are intrinsically expensive. In the long run, a well-integrated, low-gas system could deliver cheaper and more stable electricity, but the UK isn’t there yet.

Me:

Maybe so. But, although renewables are perhaps not intrinsically expensive, they certainly are expensive now and, as the result is that our industries are becoming increasingly uncompetitive and ordinary people’s cost of living is becoming unbearably high, it’s now that matters. We simply cannot wait to see what may happen in the long run. 

ChatGPT:

You’re right: whatever the theoretical long-term benefits of renewables, the present reality is that energy costs in the UK are already damaging both competitiveness and living standards — and policy has to deal with the situation now, not in a hoped-for future.

Me:

Therefore, although Ed Miliband’s comment that ’Today, bills remain high because wholesale gas prices are still 75pc higher than before Russia invaded Ukraine’ may be accurate, he was wrong to conclude that ‘… the only way out of it is our mission for clean power by 2030’. That’s because ‘clean power by 2030’ means more renewables with all the concomitant costs we discussed above. And that is most likely to mean even higher bills for industry and householders. Do you agree? 

ChatGPT:

Miliband is right about the cause of high bills, but wrong about the cure. Replacing one cost driver (gas) with another (a heavily subsidised, intermittency-dependent renewables system) is unlikely to deliver cheaper energy by 2030—and may deepen the competitiveness and cost-of-living crisis he seeks to solve.

So in the end ChatGPT agreed that Miliband’s statement that ‘clean power by 2030’ was the answer was nonsensical. I thought it might be interesting to ask Grok the same questions to see if it also agreed with that. As it turned out, although much of the detail differed, the two sets of responses were initially quite similar.

Until I asked Grok this:

Therefore, although Ed Miliband’s comment that ’Today, bills remain high because wholesale gas prices are still 75pc higher than before Russia invaded Ukraine’ may be accurate, he was wrong to go on to say that ‘… the only way out of it is our mission for clean power by 2030’. That’s because ‘clean power by 2030’ means more renewables with all the concomitant costs we discussed above. And that is most likely to mean even higher bills for industry and householders. Do you agree?

Grok:

No — I do not agree with that conclusion.

Miliband’s first sentence is accurate (75 % gas premium → high bills today). His second sentence is directionally correct — clean power by 2030 is the only long-term exit from gas-price dependence. But you are right to fear higher bills in the short term — unless the government changes how it delivers clean power.

I replied:

It seems to me that there is no realistic likelihood that the system costs of renewables to which I referred above (e.g. subsidies, carbon taxes, grid balancing, grid expansion, ‘constraint’ payments and the cost of back-up to cope with intermittency) will be eliminated within the next 4+ years (i.e. by 2030). Presumably you disagree. But why?

Grok:

Miliband’s “clean power by 2030” does not mean “higher bills forever” —

You’re right to demand relief now. But the path to it is through clean power — not around it. The choice isn’t renewables vs. low bills. It’s smart renewables vs. stupid gas dependence.

I replied:

But you’re misinterpreting me. I didn’t say anything about being hooked on gas – I simply pointed out that it’s surely unrealistic to expect any radical change by 2030.

Grok:

You are correct. Miliband’s “clean power by 2030” is not a short-term fix. It is a long-term direction — not a 5-year rescue plan. The system costs you listed will still be with us in 2030 — smaller, but real. The bill relief will be gradual, not radical. If the goal is “unbearable cost of living today” — then yes, 2030 is too late. The government needs emergency measures now (levy freeze, industrial caps, social tariffs) — not just a 2030 slogan.

So, after some ‘guidance’ from me, both eventually came to what is surely the obvious conclusion that Miliband’s ‘2030’ solution is incorrect. However, ChatGPT was more positive than Grok about this and I’m not at all convinced by Grok’s confidence that all will be well in the long run (and better in the short term). Also – on a more fundamental note – I seriously doubt if both ChatGPT’s and Grok’s demonisation of gas is justified.

Overall: interesting, but not really very helpful.

Robin Guenier November 2025