A new assured dataset the D&O desk has not used
Directors and officers underwriting in India has long relied on financials, governance disclosures and news flow. A newer source has quietly become more reliable and more material: the Business Responsibility and Sustainability Report, and specifically its assured core. BRSR Core is a subset of about thirty key performance indicators that require reasonable assurance, phased in from the top 150 listed companies in FY2023-24 to all top 1,000 listed companies by FY2026-27.
The word that matters for an underwriter is assurance. Reasonable assurance means an independent assurance provider has tested these particular KPIs to a defined standard, so they are not unverified management claims but figures someone has signed off on. That makes BRSR Core a structured, comparable, assured ESG dataset across the largest listed companies, exactly the kind of input a risk-selection process can use rather than treat as soft narrative.
The gap is that this dataset has barely been mined for D&O. The reports are long, the KPIs sit inside a larger unassured BRSR narrative, and the signal a D&O underwriter cares about, the distance between what a company claims and what its assured numbers show, is buried in text and tables. That is a problem machines are well suited to, which is the subject of this post.
Why a BRSR misstatement is a D&O trigger
Before mining the data, it is worth being precise about why an ESG misstatement is a D&O matter and not merely a reputational one. The link runs through directors' personal exposure and through the structure of a D&O policy.
On exposure, directors who sign BRSR disclosures containing material climate misstatements face personal liability under Section 447 (fraud) and Section 131 of the Companies Act 2013, and misleading disclosures can attract action under the SEBI LODR framework. So a false or materially misleading ESG statement is not a soft risk; it can put the signing directors in the frame of company-law and securities-law enforcement.
On policy structure, greenwashing-driven shareholder losses can produce claims that fall under Side C (cover for the company itself in securities claims) and Sides A and B (cover for the directors and the company's indemnification of them). That spread across the policy is what makes a BRSR misstatement a defined D&O trigger rather than a peripheral concern: the same misstatement can drive both an entity securities claim and a personal claim against the directors who signed it.
What an extraction model should look for
If the BRSR is the source and D&O the use, the question is what signals a model should pull. The useful frame is not the absolute ESG performance but the consistency between claim and assured number, because greenwashing and misstatement live in the gap between the two.
- Claim-versus-assured divergence: narrative ESG claims in the unassured BRSR text that the assured Core KPIs do not support, the company that describes itself in strong climate language while its assured emissions or intensity figures tell a flatter story.
- Year-on-year discontinuities: assured KPIs that jump or restate without explanation, which can signal either a prior misstatement or a fragile measurement process.
- Boundary and methodology shifts: quiet changes to what is counted, which can flatter a metric without any real change in performance.
- Outliers against sector peers: an assured KPI far from comparable companies, which is a flag to investigate rather than a conclusion.
A model that extracts the assured KPIs into a structured form, aligns them with the narrative claims, and surfaces these divergences turns a long document into a short list of questions for the underwriter. The output is not a verdict; it is a set of red flags a human then tests, which is the appropriate division of labour between extraction and judgement.
Where the assured-data boundary sits after March 2025
An underwriter relying on BRSR data needs to know how far the assurance actually reaches, because the boundary moved. The 28 March 2025 SEBI circular made value-chain ESG disclosure voluntary and deferred mandatory value-chain assurance. That narrows, but does not remove, the assured-data scope an underwriter can lean on.
The practical reading is that the company's own BRSR Core KPIs remain on the assured, reliable side of the line, while value-chain ESG data, covering suppliers and downstream partners, is now voluntary and not subject to mandatory assurance. For risk selection, that means an underwriter can treat the entity's own assured Core metrics as firm ground and should treat any value-chain ESG claims with more caution, because they no longer carry the same assurance backing.
This matters for the misstatement analysis. A divergence inside the assured Core KPIs is a stronger signal than a divergence involving voluntary value-chain disclosure, because the former sits within assured data and the latter does not. An extraction model should be configured to weight the two differently rather than treating all ESG numbers in the filing as equally reliable.
Getting that weighting right, and knowing how a D&O wording responds to a securities or derivative claim built on an ESG misstatement, is where the policy detail meets the data.
From red flags to a defensible underwriting view
Putting the pieces together, the workflow is a pipeline rather than a single model. Extract the assured BRSR Core KPIs into structured form. Align them with the company's narrative ESG claims. Surface the divergences, the discontinuities, the methodology shifts and the peer outliers, weighting assured-Core signals above voluntary value-chain ones. Hand the resulting red flags to an underwriter who tests them against the company's broader risk picture and its D&O history.
The value is twofold. It lets a D&O desk use a structured, assured ESG dataset that was previously too buried to mine at scale, and it grounds the greenwashing question in figures someone has assured rather than in impressions from news flow. It does not replace underwriting judgement; misstatement is a legal and factual question that a red flag only opens. But it gives the underwriter a sharper starting point and a more defensible basis for the questions asked of a risk.
The last connection is to the policy itself. A red flag only matters if the underwriter understands how the D&O wording would respond to a securities claim, a derivative action or a regulatory investigation arising from an ESG misstatement, which Side answers, which exclusions bite, and how the claim would be defended. Sarvada gives commercial insurance brokers structured, searchable access to insurer policy wordings and the intelligence around them, so the ESG signals a model surfaces can be matched against how real D&O wordings respond. Request Access to connect ESG risk selection to the cover that has to answer for it.