News from the Wire

AtData adds Gibberish Detection to ML-driven fraud detection suite

Wednesday, January 07, 2026 — 19:39:47 (UTC)

New real-time signal identifies and intercepts emails classified as synthetic, random or automated

Dedham, Mass. – AtData, a leading innovator in email address intelligence and digital trust solutions, today announced Gibberish Detection, a new machine learning-driven model in its fraud prevention suite. The capability identifies AI-generated, nonsensical, or otherwise low-intent email inputs at the point of capture. By surfacing a fast, high-confidence indicator of likely bot or synthetic addresses, the feature helps fraud and risk teams prioritize accuracy and speed while cutting operational costs.

“Stopping automated and synthetic accounts at the first touchpoint is one of the most cost-efficient ways to lower fraud exposure,” said Diarmuid Thoma, Head of Fraud & Data Strategy at AtData. “Gibberish Detection converts messy email input into a structured signal that boosts model accuracy, keeps review queues focused on real risk, and slows fraudsters instead of valuable customers.”

Across AtData’s leading activity network, approximately 5% of captured email inputs appear to be randomly generated, nonsensical, or otherwise gibberish – a baseline signal and early indicator of low-intent or automated creation. For a global on-demand services provider, AtData’s recent deployment found the gibberish detection rate doubled to nearly 10% of new orders.

Gibberish Detection analyzes the text of an email address to classify the likelihood of randomness or automation using indicators such as pattern anomalies and likely bot behavior. The resulting real-time, confidence-weighted signal strengthens identity verification, fraud screening, and risk decisioning workflows by offering:

Faster decisions, less friction: Provides an instant blocking or scoring signal to prevent low-value registrations without adding multi-step verification that annoys legitimate users.

Higher accuracy, fewer false positives: Adds a purpose-built signal for automated patterns, reducing reliance on rules that can misclassify legitimate users.

Lower operational costs: Reduces manual review volume—one of the largest expenses in fraud programs—by filtering low-value registrations at the start.

Gibberish Detection is available now through AtData’s Fraud API endpoint.

About AtData

AtData is a provider of email-centric identity and digital trust solutions that help organizations separate genuine customers from bad actors in real time. Backed by more than 25 years of historical data and models trained on billions of monthly behavioral activities, AtData delivers high-confidence signals for fraud prevention, customer acquisition hygiene, and identity resolution.

AtData Media Contact

Damian Barnes – Creative Director 770 Legacy Place, Dedham, MA 02026 dbarnes@atdata.com

Notice to readers: These are archived articles. Contact information, links and other details may be out of date. We regret any inconvenience.

Source: Company press release.

Categories: New Product