Pavesen Research Reveals AI Platforms Are Building Biographical Profiles of Wealthy Individuals From Unverified Sources
London-based advisory firm finds uncontrolled sources outnumber controlled ones by 8 to 1 in AI-generated profiles of high-net-worth individuals
LONDON, March 2026 — Pavesen, a London-based digital reputation strategy firm, has published findings from its latest research into how AI-powered search platforms represent high-net-worth individuals and business leaders. The research reveals a significant gap between how prominent individuals believe they are perceived online and the narratives AI platforms are actually constructing about them.
The study examined the information ecosystems surrounding prominent business figures and ultra-high-net-worth individuals across major AI platforms including ChatGPT, Google Gemini, and Perplexity. The consistent finding: for every authoritative source an individual controls — such as their website, LinkedIn profile, or published interviews — there are approximately eight sources outside their control, including news coverage, forum discussions, court records, and third-party commentary.
“AI platforms don’t distinguish between a verified biography and a Reddit thread,” said Tony McChrystal, Founder of Pavesen. “They treat all publicly available information as equal raw material for constructing a single, confident narrative about an individual. For people who have deliberately maintained a low digital profile, this creates the opposite of the protection they intended.”
The research also identified that individuals who had invested most heavily in traditional privacy strategies — limiting their online presence, avoiding media, and maintaining discretion — were consistently the most exposed to inaccurate or incomplete AI-generated profiles. This is because AI platforms cannot cite sources that do not exist, meaning an absence of controlled content leads to a narrative dominated by whatever uncontrolled material is available.
The findings have particular relevance for family offices and wealth advisors. Deloitte’s 2024 Family Office Cybersecurity Report found that 43 per cent of family offices had experienced a cyberattack within 24 months, with those managing over $1 billion significantly more targeted. Pavesen’s research suggests that the reputational dimension of these incidents — as amplified and preserved by AI — is rarely accounted for in existing risk frameworks.
“What we’re seeing is a new category of exposure that most advisory firms haven’t mapped yet,” McChrystal added. “Investment oversight, cybersecurity, and succession planning all sit within formal governance structures. Reputation, in most family offices, does not — yet it’s increasingly the first thing a counterparty, journalist, or regulator encounters when researching an individual.”
Gartner has projected that by 2028, 50 per cent or more of internet users will rely on AI-generated answers rather than traditional search engines, suggesting the scale of this issue will continue to grow.
About Pavesen
Pavesen advises high-profile individuals, families, and leadership teams on how they are represented across search engines, AI platforms, and the wider digital environment. The firm is typically engaged during periods of heightened exposure — scrutiny, transition, dispute, transaction, or reputational threat — when digital interpretation can materially influence judgement. More information is available at www.pavesen.com
Media Contact:
Tony McChrystal
Founder, Pavesen
info@pavesen.com
+44 333 050 3125