Your corporate communications and annual report are being read by systems you cannot see.

 

AI systems are already interpreting your corporate communications. They are forming judgements, generating summaries, and making inferences at scale. Jones+Palmer’s GLASSBOX shows you what they see, before the market acts on it.

The Problem: AI can feel like a black box between your corporate communications and your reputation.

We built Glassbox so you can see what AI sees, before the market does.

Every analyst briefing, every governance statement, every strategic narrative in your annual report is now source material for AI systems, from investor research tools to regulatory screening engines. But you can't see inside the process.

Most don't know what an LLM concludes when it reads your Chair's statement, how it summarises your strategy, or what connections it invents between your risk disclosures.

That's the black box.

When an LLM infers something you didn't say, or invents connections you did not make, the reputational risk still lands on you. And it's harder to detect, harder to correct, and harder to explain to the board than a misquote.

Glassbox-visual.jpg image

What Glassbox does and aligns with

As AI systems increasingly shape stakeholder understanding, GLASSBOX aligns with the FRC’s 2026 guidance by helping organisations stress-test communications before AI interpretation becomes market perception. Combining AI analysis with human expertise, GLASSBOX can also assist organisations identify risks, understand interpretation across sector and company-specific issues, and strengthen how narratives, disclosures and financial messaging are understood.

Inference audit

See what an LLM concludes from your report, the gaps it fills, the claims it invents, the connections it assumes.

Hallucination mapping

Identify where your language creates ambiguity that AI systems resolve incorrectly and where that creates material risk.

Narrative resilience score

A clear, defensible measure of how accurately your corporate narrative survives AI interpretation.

Remediation guidance

Expert-validated recommendations to close the gaps. Not AI-generated fixes, human-led editorial counsel.

ChatGPT Image May 13, 2026, 10_01_29 AM.png image

AI-assisted research is already changing how companies are discovered, summarised and assessed.

Your reporting team proofs for human readers. Your legal team checks for regulatory compliance. Who is checking what an LLM concludes when it reads your Chair's statement?

GLASSBOX fills that gap, sitting between your communications team and the AI systems that now mediate how your story reaches the market.

Built for

Co-Sec

Governance Assurance

Reporting Team

Narrative Integrity

IR

Market confidence

ChatGPT Image May 13, 2026, 09_58_08 AM.png image

How it works

Three steps. Full transparency. No black boxes.

Step 1

Upload

Upload your annual report or simply your company name.

Step 2

Stress-test

Multiple independent AI interpretation layers test how large language models retrieve, summarise, infer and draw conclusions from your communications.

Step 3

Assurance report

Receive three independent reports and one consolidated view showing where inference, ambiguity and hallucination risk matter most, alongside expert remediation guidance.

See what AI sees in your report

Request a confidential GLASSBOX assessment for your next corporate communications cycle.

Request an assessment

J&P HQ Shoot BM 019 BIRO image