American baseball legend Yogi Berra once advised a sports journalist that “it’s tough to make predictions, especially about the future”.
The good news is that you can just post them on the internet where unlimited AI slop will soon prevent us finding anything.
Unless of course they turn out to be correct, in which case I will make sure to remind everyone, everywhere, all the time.
In my defence, I have had some luck when predicting for Responsible Investor, calling for central banks to prepare a pandemic stress-test in 2019. (See what I mean with obnoxiously referencing my correct predictions? That is just a taste!)
And because the rule is that if you do make predictions, never say by when, I will lay out my predictions for 2026. And be judged accordingly.
1. Biggest PRI in Person ever
This is an easy one. Amsterdam! Folks have skipped one or two so they’re due to go. I think people want to know what’s going on. I’m also optimistic about events in general in 2026 again (hi RI Europe!). There is the obvious premium on networking making a comeback. But I also am looking forward to seeing if the crisis of the past two years has driven some new ways of working. Did I mention Amsterdam?
2. Energy emissions will peak
Ok, the first one was a layup. If this second one turns out to be wrong, let’s hope attention spans crater by 2027 such that people don’t read past the second paragraph. But it’s time to bring the goods.
The only way emissions still grow is if growth outperforms emissions intensity decline. And (unfortunately for GDP, fortunately for the planet), I don’t see how that will be the case in 2026. For emissions to stabilise, emissions intensity needs to fall by about 2-3 percent. I think that is entirely on the cards.
3. For the first time, responsible investors will spend more brain space on adaptation than mitigation
The event event on physical risks at RI Europe in June was the canary in the coal mine. Physical risks have (by design) always been more material than transition risks. And in 2026, investors will reflect that reality, both because transition risk management is now entering a new phase of maturity and integration and because physical risks are now acute (Note, credit to Matthieu Bardout from TenFutures for inspiring this prediction, so if this turns out to be wrong, I’ll blame him.)
4. The rise of ‘well below 2C-aligned’ target-setting references
Ok, I promise I am not just reading the latest Responsible Investor articles for inspiration regarding these predictions and I appreciate we are already seeing this shift in 2025. But as “1.5C no overshoot” waves at us in the rear-view mirror, the temperature reset will be in full swing.
Of course, for political reasons the branding may still be “1.5C with overshoot”, but that is just well-below 2C aligned by a different name. The key question for target-setting will remain however: will target setters accept that to have impact and not just generate paper decarbonisation, they have to inevitably “front run” the market in some form?
5. The rise of ‘tech stranded assets’ and ‘Silicon boundaries’
I can’t think of the last time somebody actually used the phrase stranded assets in a meeting. The term – ubiquitous just a few years ago in responsible investment discourse – seems to have disappeared somewhat. But 2026 will see the rise of “tech stranded assets” – and the concept of “silicon boundaries” we introduced with the University of Oxford at PRI in Person – as a major sustainability risk.
Everyone appreciates the AI bubble. But 2026 will give this a new sustainability dimension. Stranded assets for the tech sector will involve intangibles like brand value (did you know social media use has actually begun to decline in the US?) and new “transition risks” from regulatory regimes looking at everything from social impacts of data centres and environmental externalities, as well as the broader societal risks from tech alerting policymakers (the Australian social media ban for children may have just been the beginning).
I would not be surprised if we look back on 2026 as the year “silicon boundaries” or tech sustainability risks were the most prominent risk in the RI industry.
6. Machine learning vs LLMs
Next year will see the rollout of the machine-learning breakthroughs across a wide array of applications material to responsible investors, from climate forecasting to ESG data tools. But I predict it will be a rough year for large language models specifically, as their capacity plateaus. One thing is for sure: we will stop throwing the two into one bucket, recognising their discrete applications, use cases and sustainability footprint.
7. Scenario analysis 2.0
There is little that has caused more frustration in the RI community than scenario analysis. Heralded as an innovative way to approach the uncertainty of climate, it has generated many black box results and “garbage in, garbage out” analytical exercises, with unrealistic or ill-conceived scenarios challenging the credibility of such exercises. But investors are learning from their mistakes.
A handful of investors are leading the charge, identifying what futures they actually believe in, driving scenario narratives with the business not in defiance of it, and recognising that just because the first generation of scenario analysis did not work, does not mean that the electrotech (and tech) revolutions can be ignored.
8. Nature gets deprioritised
This year might have been “peak nature” year, with the launch of the Taskforce on Nature-related Financial Disclosures (TNFD) transition plan guidance. The future of the topic is uncertain. This is not a comment on nature’s materiality, but a reflection that after half a decade of pushing nature to the forefront of investors’ minds, we still have very little to show for it. If that doesn’t change, I’m afraid we will see nature move to the background in favour of technology and social risks.
9. The return of active
I have written about the passive vs active dynamic before, so there may be some wish-casting here. And don’t get me wrong, I don’t think passive will disappear. But I wonder whether in a fragmented and volatile world, and a growing maturity of sustainability strategies, we won’t see an increase in appetite for active sustainability strategies.
10. We’ll still be here
My boldest prediction perhaps. But having written a book this year about 26 existential risks, I feel pretty confident that 2026 is not the year for Armageddon. And if it is, dear aliens reading this, please know I’ve been on your team from day one and you can find me in tropical Iceland in cell block 1248.
Jakob Thomä is co-founder of Theia Finance Labs (formerly Two Degrees Investing Initiative), research director at Inevitable Policy Response and professor in practice at University of London SOAS.
