Sensity
Founded 2018 · 26 employees on LinkedIn · 4 known investors
Sensity AI develops deepfake detection technology that identifies synthetically generated or manipulated images, videos, and audio by using deep learning to detect pixel-level and acoustic artifacts left by generative AI. Its forensically robust analysis is aimed at states, organizations, businesses, individuals, and law enforcement handling cases of fraud, harassment, and abuse.
Also known as Deeptrace · Sensity AI
Founders & leadership
Sensity was founded in 2018 by Giorgio Patrini.
Investors · 4
Company profile
researched Aug 2026Sensity AI develops software that detects AI-generated and manipulated media across images, video and audio. Its platform applies a multilayer analysis engine that combines pixel-level visual forensics, raw audio-spectrum analysis for synthesized or cloned voices, and inspection of file container signatures such as codecs, timestamps and metadata, alongside behavioral cues and cross-modal inconsistency checks. Users can submit files or URLs and receive an assessment within seconds.
Rather than returning a binary verdict, the product generates forensic reports containing confidence scores, visual indicators, explainability tools, audit trails and structured documentation, positioned for use as evidence in corporate investigations, law enforcement cases and court proceedings. The company also operates a threat intelligence unit that identifies malicious deepfake content targeting specific individuals, organizations or states as it appears online. The system is offered as a cloud-based service and as an on-premise deployment with fully offline operation and GPU-accelerated inference, aimed at government agencies and financial institutions with data-sovereignty requirements.
Sensity AI describes application areas including digital forensic evidence and media authentication, disinformation and influence operations, indecent images of children (IIOC/CSAM), non-consensual intimate imagery and other sexual crimes, and deepfake-enabled fraud such as impersonation of trusted figures and synthetic identities in KYC and contact-center workflows.
Founding story
Co-founder Giorgio Patrini was a postdoctoral researcher at the University of Amsterdam in 2017, working on deep generative models with Max Welling, including image generation with GANs and VAEs. Anticipating that generative tools would enable coercion, fraud and attacks, he trained what the company describes as possibly the first deepfake detector in March 2018 and concluded that forensically robust detection able to withstand courtroom explainability standards would be needed. He joined with Francesco Cavalli, a threat intelligence specialist researching fake news and its effects on internet communities and politics, and the company was founded late in 2018.
Business model
Sensity AI sells its deepfake detection software to organizations, offering a web application, API and SDK, with both cloud-based and on-premise deployment options. Prospective customers can start a trial or contact sales.
Traction
The company reports 98% accuracy on public datasets, more than 900,000 incidents identified in 2025, an average of 35 minutes saved on manual review per case, and deployment in more than 30 countries. It cites deployments inside high-security forensic environments, intelligence and security teams, and on-premise banking environments, and inclusion in the NIST Computer Forensics Tools & Techniques Catalog.
Latest developments
As of early 2026 the company publishes threat intelligence on cognitive warfare and narrative-based influence operations, and notes coverage of its work in outlets including The Cipher Brief and the Kyiv Post, as well as its listing in the NIST Computer Forensics Tools & Techniques Catalog.
▸Full profile — market position, technology, go-to-market, geography, history
Market position
The company positions itself as the world's first deepfake detection company and a first mover since 2018, and states that it has won public and private tenders against competitors that raised substantially more venture capital. It differentiates on forensic robustness rather than binary classification.
Sensity AI emphasizes multilayer detection across visual, audio, file-structure and metadata signals rather than single-layer classification; forensically robust, explainable and reproducible outputs designed to meet evidentiary standards; on-premise and offline deployment for organizations with strict data policies; and pairing detection with an in-house threat intelligence capability.
Technology
The detection engine is trained on large sets of known deepfake data using deep learning to identify the 'digital fingerprints' left by synthetic generation or manipulation, at the pixel level for visual media and through subtle acoustic artifacts for audio. Separate layers analyze the raw audio spectrum for signs of voice synthesis and cloning, and the technical signatures inside file containers, including codecs and timestamps. The engine stacks these independent forensic signals and adds metadata, behavioral and cross-modal inconsistency checks. Outputs emphasize explainability and reproducibility, producing traceable reports intended to be verifiable in court. Deployment supports full offline operation and GPU-accelerated inference, and integration into existing evidence-ingestion pipelines, identity verification systems and call-center systems.
Go-to-market
Direct enterprise and public-sector sales, including participation in public and private tenders, supported by free trials, expert consultations, developer documentation, published threat intelligence articles and downloadable reports aimed at law enforcement professionals and forensic analysts.
Government and judicial authorities, intelligence and national security teams, law enforcement and digital forensic units, banks and financial institutions handling identity verification, KYC and contact-center workflows, private enterprises, and individuals affected by malicious deepfakes.
Geography
Deployed in more than 30 countries; the company's research origins trace to the University of Amsterdam.
History
The company dates its work in AI-generated deepfake detection to 2018, describing itself as a first mover in the field. Its stated approach combines deep learning with classical image, video and audio forensics and situates detection within threat intelligence. Over time it has been deployed inside high-security forensic environments, intelligence and security teams, and on-premise within banking environments. As of early 2026 it reports deployments in more than 30 countries and lists itself in the NIST Computer Forensics Tools & Techniques Catalog.
Compiled by commissioned research from 6 cited public sources — announcements, filings, and press listed under research sources below.
Key figures
latest reportedCompany-reported or press-reported figures, each dated to when it was claimed — not independently audited.
Competitors · 5
by search overlap
Companies competing with Sensity for the same Google search keywords, organic and paid, via search-intersection analysis.
Timeline · 1
launches, deals, and filingsSensity AI is listed in the NIST Computer Forensics Tools & Techniques Catalog (toolcatalog.nist.gov).
Dated company events from announcements, filings, and press; legal rows summarize public dockets and regulator releases.
In the news
▸Research sources · 6
primary sources listed
- Sensitysensity.ai · web
6 public sources were cited for this profile; the first-party ones are listed here.
Frequently asked questions
- What does Sensity do?
- Sensity AI builds forensic-grade deepfake detection for images, video and audio used by law enforcement, governments and banks.
- Who founded Sensity?
- Sensity was founded by Giorgio Patrini in 2018.
- Who are Sensity's investors?
- Sensity's investors include Betaworks, GMG Ventures, Mercuri.vc, Betaworks Ventures Management, Llc.