Skip to content

InfoHawk

InfoHawk is an AI platform that maps online scam and fraud infrastructure to detect and block deception in real time.

Visit Website ↗ + Add to Compare Claim This Company
35/100Emerging / Unranked

Overview

InfoHawk is an AI-native fraud-and-deception detection platform founded by Rob Leathern (two decades in platform integrity roles at Meta and elsewhere), Ben Poiesz (former Google Android product leader and Grammarly cloud and AI infrastructure lead), and Jamie McCrindle (Meta engineer focused on fake-account prevention). The company closed a $2.25 million pre-seed round on June 8, 2026, led by Moonshots Capital, with angel participation from former FTC Chairman Jon Leibowitz, AppNexus founder Brian O’Kelley, former Meta Ads head Rob Goldman, and GitHub CTO Vlad Fedorov. InfoHawk is headquartered in Austin, Texas.

The product evaluates URLs, ads, accounts, and other digital assets across text, image, video, and infrastructure signals (domains, IPs, phone numbers, payment rails), correlating them to identify coordinated scam and deception campaigns rather than scoring isolated pieces of content. It returns a risk verdict in under 300 milliseconds via a REST API, and also offers an MCP server for AI agents and a no-code ‘Agent Investigator’ workspace.

InfoHawk’s stated differentiator is mapping the underlying fraud infrastructure rather than just scanning content, to catch coordinated campaigns and evasion attempts that content-only scanners miss. The company is only months past its first funding round, with no independently verified efficacy data or named enterprise customers public yet, and its primary use case of consumer scam and trust-and-safety protection is a partial rather than core fit for a CISO-centered enterprise security matrix.

Innovation Matrix Assessment

Innovation Velocity 5/10

In the months since forming, the team shipped a working multimodal detection engine with sub-300ms verdicts, an API, and an MCP server, but there is no track record yet of repeatable product advances.

Operational Value 4/10

Useful for trust-and-safety and fraud teams to block scam infrastructure in real time, but it serves consumer-fraud and brand-protection functions more than core CISO security operations.

Market Momentum 2/10

Only a single $2.25M pre-seed round (June 2026) is publicly disclosed; no named enterprise customers, pilots, or partnerships were found.

Category Disruption 3/10

Multimodal fraud and scam detection is a crowded space (Sift, Arkose Labs, DataDome, HUMAN Security); InfoHawk's infrastructure-mapping angle is a reasonable differentiator but not evidence of category redefinition yet.

Real-World Efficacy 2/10

No independent testing, named scam-network takedown, or customer case study was found; all efficacy and speed claims trace back to the company's own site and investor commentary.

Enduring Relevance 5/10

AI-generated scams and deepfake-enabled fraud are a growing, durable threat category, so the problem space will matter in 3-5 years even though this specific vendor's staying power is unproven.

Why CISOs Should Care

For CISOs whose remit includes brand protection, fraud, or consumer trust-and-safety, InfoHawk offers a way to detect and block AI-generated scam infrastructure before it reaches customers, in near real time.

What Makes It Different

Rather than scoring individual pieces of content, InfoHawk maps the underlying fraud infrastructure to catch coordinated scam campaigns and evasion attempts across platforms and customers.

The Matrix Verdict

35/100 — EMERGING / UNRANKED

InfoHawk is a real, credibly-backed pre-seed company with an experienced trust-and-safety founding team, but it is only months past its first funding round with no disclosed enterprise customers or independent efficacy evidence, and its consumer scam and fraud focus is a partial rather than core fit for an enterprise CISO-focused matrix.

Editorial Note: Claims vs. Verified Findings

Speed and cost claims come solely from InfoHawk's own site and investor commentary; no independent benchmark, named customer, or incident takedown was found to corroborate them.

Sources