CONSOB, the Italian financial markets regulator, and the University of Trento have jointly developed a prototype for detecting potentially misleading messages published by bond issuers.
The prototype uses large language models, specifically Google’s domain-adapted ClimateBERT and ESGBERT models, in combination with a proprietary dictionary to map a defined set of
Register for free to keep reading.
To continue reading this article and unlock full access to GRIP, register now. You’ll enjoy free access to all content until our subscription service launches in early 2026.
- Unlimited access to industry insights
- Stay on top of key rules and regulatory changes with our Rules Navigator
- Ad-free experience with no distractions
- Regular podcasts from trusted external experts
- Fresh compliance and regulatory content every day