Bold Security deploys on-device AI to guard enterprise data from shadow AI
Bold Security’s CEO Nati Hazut explains that the shift of work from browsers to local devices has created new data‑exfiltration channels that traditional network‑based security tools miss. By running small language models directly on laptops and desktops, Bold can classify sensitive information in real time, coach users, redirect activity, or block risky actions across browsers, desktop apps,…
Key points
- Bold Security runs small language models on devices to classify and block sensitive data in real time
- Endpoint‑first approach sees activity that network, CASB, and DSPM tools miss, including autonomous AI agents
- Models use about 20% of the compute of legacy non‑AI agents, keeping user experience smooth
Hazut says the endpoint‑first approach lets the system understand the context and intent behind data movements, reducing false positives that plague rule‑based DLP. The models are engineered to use about 20% of the compute resources of legacy non‑AI agents, keeping laptops responsive while providing latency‑free decisions. He notes that autonomous agents can act at machine speed, so security must operate equally fast, distinguishing between legitimate employee use and agent‑driven risks.
Looking ahead, Hazut predicts every device will become an intelligent security layer that continuously decides which humans and agents may interact with particular data, giving enterprises granular, real‑time control over AI‑driven workflows.
Nati Hazut, CEO and Co-Founder of Bold Security
Unite.AI · 23 October 2026
Loading the full article…
This text was published by Unite.AI and written by Antoine Tardif, CEO & Founder of Unite.AI. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Enterprise & Industry
All →- RAVEL CEO Denise Muyco says AI factories need orchestration, not just more GPUs · 1 src
- Morgan Stanley blocks third‑party AI agents after banks detect unauthorized usage · 6 src
- Opinion: security teams should add a deterministic second opinion to AI scans · 1 src
- Amazon scrubs plans to build four data centers in Loudoun County · 1 src
- IBM CEO says businesses prefer open-weight models to costly frontier AI · 1 src
Comments
via GitHub Discussions