Menu
Sign In Search Podcasts Charts People & Topics Add Podcast API Blog Pricing
Podcast Image

LessWrong (Curated & Popular)

"AI found 12 of 12 OpenSSL zero-days (while curl cancelled its bug bounty)" by Stanislav Fort

28 Jan 2026

Transcription

Chapter 1: What recent vulnerabilities were discovered in OpenSSL?

0.031 - 5.496 Stanislav Fort

AI found 12 of 12 OpenSSL zero days, while Curl cancelled its bug bounty.

0

6.437 - 10.602 Unknown

By Stanislav Fort. Published on January 27, 2026.

0

10.882 - 37.168 Stanislav Fort

This is a partial follow-up to I'll Discovered three new OpenSSL vulnerabilities from October 2025. TLDR. OpenSSL is among the most scrutinized and audited cryptographic libraries on the planet, underpinning encryption for most of the Internet. They just announced 12 new zero-day vulnerabilities, meaning previously unknown to maintainers at time of disclosure.

0

38.209 - 57.392 Stanislav Fort

We at I'll discovered all 12 using our AI system. This is a historically unusual count and the first real-world demonstration of AI-based cybersecurity at this scale. Meanwhile, Curl just cancelled its bug bounty program due to a flood of AI-generated spam, even as we reported five genuine CVEs to them.

0

Chapter 2: How did AI contribute to the discovery of these vulnerabilities?

57.811 - 78.287 Stanislav Fort

AI is simultaneously collapsing the median, slop, and raising the ceiling real zero days in critical infrastructure. Heading Background We at ILE have been building an automated AI system for deep cybersecurity discovery and remediation, sometimes operating in bug bounties under the pseudonym Giant and Eater.

0

79.348 - 98.01 Stanislav Fort

Our goal was to turn what used to be an elite, artisanal hacker craft into a repeatable industrial process. We do this to secure the software infrastructure of human civilization before strong AI systems become ubiquitous. Prisaically, we want to make sure we don't get hacked into oblivion the moment they come online.

0

98.513 - 108.726 Stanislav Fort

No reliable cybersecurity benchmark reaching the desired performance level exists yet. We therefore decided to test the performance of our AI system against live targets.

0

109.364 - 126.066 Stanislav Fort

The clear benefit of this is that for a new, zero-day security vulnerability to be accepted as meriting a CVE, a unique vulnerability identifier, it has to pass an extremely stringent judgment by the long-term maintainers and security team of the project, who are working under many incentives not to do so.

0

126.106 - 134.518 Stanislav Fort

Beyond just finding bugs, the issue must fit within the project's security posture, that is what they consider important enough to warrant a CVE.

Chapter 3: What challenges did the Curl bug bounty program face?

135.659 - 159.026 Stanislav Fort

OpenSSL is famously conservative here. many reported issues are fixed quietly or rejected entirely. Therefore our benchmark was completely external to us and in some cases intellectually adversarial. We chose to focus on some of the most well-audited, secure, and heavily tested pillars of the world software ecosystem. Among them, OpenSSL stands out.

0

160.168 - 170.184 Stanislav Fort

Industry estimates suggest that at least two-thirds of the world's internet traffic is encrypted using OpenSSL and a single zero-day vulnerability in it can define a security researcher's career.

0

171.245 - 179.618 Unknown

It is a very hard target to find real, valuable security issues in. Heading Fall 2025

0

179.598 - 193.194 Stanislav Fort

Our first OpenSSL results. In late summer 2025, six months into starting our research, we tested our AI system against OpenSSL and found a number of real, previously unknown security issues.

0

Chapter 4: What is the significance of the 12 zero-day vulnerabilities found?

194.295 - 224.525 Stanislav Fort

In the full 2025 OpenSSL security release, four CVEs in total were announced from 2025 of the format. There's a code block here in the text, out of which three were found, responsibly disclosed, and in some cases even fixed by us, or more precisely by our AI system. You can read more in our original blog post. Specifically, these were two moderate severity issues. CVE-2025-9230.

0

224.985 - 246.092 Stanislav Fort

Out-of-bounds read-write in the RFC3211KEK unwrap operation for CMS password-based encryption, potentially leading to memory corruption or code execution. This bug had been present since 2009, undetected for over 15 years. CVE 2025-9231.

0

246.673 - 269.27 Stanislav Fort

Timing side channel in SM2 elliptic curve signatures on 64-bit ARM, where variations in execution time during modular arithmetic could in principle allow private key recovery through careful remote observation. This is a subtle, logic-level vulnerability where the correctness of the code obscured a timing leak that only emerged under specific hardware conditions.

0

Chapter 5: How does AI influence the future of cybersecurity?

270.371 - 289.39 Stanislav Fort

We also found a single low-severity CVE. CVE 2025-9232. Out-of-bounds read in HTTP client. There's a code block here in the text. Handling when parsing IPv6 hosts, triggering a controlled crash.

0

289.657 - 312.306 Stanislav Fort

Independently, the Frontier of the Year 2025 forecasting project by Gavin Leach, Lauren Gilbert, and Ulka Ageva looked out for AI-driven vulnerability discovery in critical infrastructure as one of the top AI breakthroughs of 2025, assigning it a 0.9 probability of generalizing and placing it at number three overall by expected impact, resolving as, quote,

0

313.568 - 326.622 Stanislav Fort

Google's big sleep agent and the startup aisle found dozens of critical vulnerabilities in some of the main infrastructure of the internet. Linux, Curl, OpenSSL, and SQLite. Frontier of the year 2025.

0

326.782 - 328.984 Unknown

End quote.

0

Chapter 6: What are the implications of AI in vulnerability discovery?

329.865 - 349.11 Stanislav Fort

For context on our approach, our system handles the full loop equals scanning, analysis, triage, exploit construction, if needed and possible, patch generation, and patch verification. Humans choose targets and act as high-level pilots overseeing and improving the system, but don't perform the vulnerability discovery.

0

350.112 - 378.253 Stanislav Fort

On high-profile targets, we additionally review the resulting fixes and disclosures manually to ensure quality, although this only rarely changes anything. Heading. January 2026. 12 out of 12 new vulnerabilities. Just today, January 27, 2026, OpenSSL announced a new security patch release, publishing 12 new zero-day vulnerabilities, including a very rare high-severity one.

0

378.875 - 399.868 Stanislav Fort

Of the 12 announced, we at Al discovered every single one of them using our AI system. One vulnerability, CVE-2025-11187, was also co-reported by a security researcher Hamza from Metadust 33 days after our initial disclosure. Congratulations on representing humanity in this virtuous race.

0

Chapter 7: How do OpenSSL maintainers respond to AI-generated findings?

400.99 - 405.64 Unknown

Party popper! Out of the 12 new CVEs, 10 were assigned.

0

406.682 - 432.734 Stanislav Fort

There's a code block here in the text. Identifiers and 2 already belong to the year 2026 with. There's a code block here in the text. S adding this to the 3 out of. 4. CVEs we already had in 2025 previously, this means that ILE, and by extension AI in general, is responsible for discovering 13 out of 14 zero-day vulnerabilities in OpenSSL in 2025.

0

433.816 - 445.93 Stanislav Fort

Both the count and the relative proportion have been increasing as a function of time and are overall historically very atypical. The 12 vulnerabilities span a significant breadth of OpenSSL's code base.

0

Chapter 8: What does the future hold for AI in cybersecurity?

445.95 - 451.5 Unknown

Here they are sorted by severity. Subheading. High severity, 1.

0

452.602 - 480.938 Stanislav Fort

CVE 2025-15467. Stack buffer overflow in CMS auth enveloped data parsing. The overflow occurs prior to any cryptographic verification, meaning no valid key material is required to trigger it, making it potentially remotely exploitable against any application parsing untrusted CMS content. For context, high severity or above CVEs in OpenSSL have historically averaged less than 1 per year.

0

481.981 - 485.006 Unknown

Subheading. Moderate severity, 1.

0

486.128 - 497.89 Stanislav Fort

CVE 2025-11187. Stack buffer overflow and null pointer dereference in PBM AC1 parameter validation during PKCS number 12 MAC verification.

0

498.951 - 506.972 Unknown

Co-reported by Hamza from Metadust 33 days after our disclosure. Subheading. Low severity, 10.

508.095 - 549.552 Stanislav Fort

CVE for 2025-15 for 68, CVE for 2025-15 for 69, CVE for 2025-66199. CVE 2025-68160 CVE 2025-69418 CVE 2025-69419 CVE 2025-69420 CVE 2025-69421 CVE 2026-22795 CVE 2026-22796 Listed primarily for completeness sake.

549.532 - 578.205 Stanislav Fort

These span QUIC, PKCS number 12, PKCS number 7, CMS, TLS 1.3, and BIO subsystems, including heap overflows, type confusions, null dereferences, and a cryptographic bug where OCB mode leaves trailing bytes unencrypted and unauthenticated. Three of these bugs date even back to 1998-2000, having lurked undetected for 25-27 years.

578.287 - 606.1 Stanislav Fort

One of them, CVER 2026-22796, predates OpenSSL itself and was inherited from SS Lee, Eric Young's original SSL implementation from the 1990s. Yet it remained undetected by the heavy human and machine scrutiny over the quarter century. Even at low severity CVE is a higher bar than might be obvious. The vast majority of reported issues don't qualify as security vulnerabilities at all.

606.94 - 628.982 Stanislav Fort

Of those that do, most are bugs that get fixed without CVEs as standard PRS. To receive a CVE from OpenSSL, an issue must pass their conservative security posture and be deemed important enough to track formally. Low severity in OpenSSL still means a real, externally validated security vulnerability in well-audited critical infrastructure.

Comments

There are no comments yet.

Please log in to write the first comment.