Anyday Security Launches Free AI Exploit Platform

ExploitTotal.ai offers more than 200 AI-generated exploits, most for vulnerabilities with no known public proof of concept.

Updated on Nov. 21, 2026 in Cybersecurity

Isometric editorial illustration of a steel padlock with its shackle sprung open and a hairline crack across its body, set on a flat cream surface, evoking a breached security barrier.
Anyday Security launched ExploitTotal.ai, a free platform offering more than 200 AI-generated exploits for patched vulnerabilities, most with no known public proof of concept. AI Illustration. Upload story photo >

Anyday Security, a New York start-up with an R&D center in Tel Aviv, launched ExploitTotal.ai, a free platform providing AI-generated exploits for patched vulnerabilities. The company says the exploits were produced by GLM-5.3 from known vulnerabilities, with more than 70% targeting flaws that have no known public proof of concept.

Why it matters

Anyday Security says defenders need to test their exposure and build detection rules before attackers exploit patched vulnerabilities. The launch comes as organizations have historically been slow to remediate actively exploited flaws.

Anyday Security checked successful exploits against the U.S. National Vulnerability Database, CISA's known exploited vulnerabilities catalog and public exploit repositories. In one demonstration, the model built a full-chain browser exploit in minutes for less than a dollar.

The players

Anyday Security

A start-up based in New York with an R&D center in Tel Aviv that builds tools around AI-driven vulnerability research and exploit generation.

GLM-5.3

The AI model used by Anyday Security to generate exploits from known vulnerabilities and their public patch descriptions.

The details

Anyday Security gave GLM-5.3 more than 200 known vulnerabilities together with only the vendor's public description and the patch. Each exploit on ExploitTotal.ai is graded on a five-level scale, from crashing the target program to taking full control of it.

Timeline

  1. Microsoft fixed 1,411 vulnerabilities in its October 2026 Patch Tuesday.

  2. Anyday Security launched ExploitTotal.ai on November 21, 2026.

The Tech Race

The launch extends the remediation gap documented in Verizon's 2026 Data Breach Investigations Report, where organizations took a median of 43 days to fix actively exploited flaws and fully fixed only 26%. It shows AI tools compressing the time between a vendor patch and a working exploit.

Security teams can use the publicly available catalog to test exposure to patched vulnerabilities and build detection rules before attackers adapt. Organizations that delay applying vendor patches face a higher chance that a working exploit already exists.

The takeaway

Patch deployment is now a race against AI-assisted exploit generation rather than a routine maintenance task. Teams should prioritize testing for patched flaws that have no public exploit yet, since those are the ones most likely to be weaponized quickly.

Further reading

For more on the threats behind this launch, see Cybersecurity coverage.

More information

Explore the free ExploitTotal.ai vulnerability and exploit catalog.

Source note: This article includes information reported by Cybersecurity Insiders.