AI in ESG Strategy: Why Human Oversight is Non-Negotiable
AI tools are increasingly being used by companies to manage their ESG performance. From drafting policies to automating reporting, they can bring speed, efficiency, and scale.
For mid-sized businesses - often without in-house sustainability expertise - that acceleration can make a real difference.
What we find with our clients, however, is that while AI is excellent at getting them started and performing certain tasks, it rarely gets them all the way to where they need to be.
Why? Because ESG isn’t just about producing documents. It’s about demonstrating credible performance in the eyes of your customers, investors, and regulators. And credibility requires context, judgement, and accountability - aspects that can’t be automated.
Where AI adds most value in ESG
Used responsibly, AI can be a powerful accelerator:
Drafting the first version of a policy, Net Zero plan, report or questionnaire response.
Automating parts of the reporting process.
Researching and synthesising frameworks and regulations into usable summaries.
Identifying patterns in data that could take a human team days to find.
For stretched ESG or leadership teams, this can free up valuable time.
Where AI currently falls short
What we see (and hear from procurement teams) is that AI-generated ESG content lacks the context and detail that decision-makers look for. It often:
Misses the sector-specific risks or priorities that matter most.
Produces generic language that undermines credibility in tenders or investor discussions.
Risks over-promising or creating inconsistencies that open the door to accusations of greenwash.
Risks misguided action or reporting based on in-correct AI ‘hallucinations’, especially in the fast-evolving world of ESG
This isn’t theoretical.
We know from conversations with public sector procurement teams, that they can quickly spot bland AI-generated responses and when they do, they either seek detail and clarification or, worse, give the supplier a low mark and exclude them from the tender process. Suppliers are reporting lost contracts, missed investment opportunities, and reputational damage as a result.
The sustainability cost of AI
There’s also the question of footprint. Training and running AI models consumes significant energy and water. For companies serious about reducing emissions, relying too heavily on AI without working out the most efficient way to do so risks undermining the very ESG commitments they’re trying to demonstrate.
The solution isn’t to avoid AI, but to use it with purposeand efficiency: automation for speed, paired with oversight for accuracy and integrity.
Why human oversight matters
Human expertise provides what AI cannot:
Relevance - tailoring ESG to what customers and investors truly care about.
Accuracy - ensuring claims and data stand up to scrutiny.
Confidence - giving boards, regulators, and stakeholders assurance that commitments are credible.
With the right oversight, AI becomes a genuine enhancer - speeding up the process while maintaining integrity in the output.
Thrive’s perspective
At Thrive, we see AI as a tool, not a replacement. What our clients need is the blend: the acceleration that AI brings, grounded in the judgement only people can provide.
Our Ask Thrive service is designed to work alongside your AI-generated work.
It gives companies quick, expert oversight to ensure their ESG work is credible, relevant, and ready to stand up under scrutiny.
Because when it comes to ESG, it’s not about producing more policies. It’s about producing commitments you can deliver on, to build confidence and trust with those in your business ecosystem.