Tim Michaud (@TimGMichaud) profile picture

Founder @ New thing - (YC Alum) still a Security Nerd.

Middle of Freakin Nowhere
Joined March 2012
Photo by Tim Michaud (@TimGMichaud)Photo by Tim Michaud (@TimGMichaud)Photo by Tim Michaud (@TimGMichaud)Photo by Tim Michaud (@TimGMichaud)Photo by Tim Michaud (@TimGMichaud)Photo by Tim Michaud (@TimGMichaud)Photo by Tim Michaud (@TimGMichaud)Photo by Tim Michaud (@TimGMichaud)Photo by Tim Michaud (@TimGMichaud)Photo by Tim Michaud (@TimGMichaud)
b4c0d6f4c83c492d4a73e29e73deeb3bbfadbbc56fb3497d38a6715de75cadf2 Vagueposting about tailscale is what made me find this bug (which was somehow missed by the AI they used?) is hilarious to me
1
3
244
Similar concerns to Solnik, but exdev is also niche and I have yet to talk to a non FAANG who care about exdev as a higher priority problem to solve vs. "Help me get through all of this bullshit llms are pumping out"
this is the most underrated response in cyber security and AI combined to this date. Im really whiling to read your thoughts on why I am so delusional to provide CyberKimi at a $4/M based on everything that we published so far (ACE'd few Google Chrome ndays full chains on ExploitBench and on real world patch-diffs, almost ACEd (97%) CyberGym, beat GLM and ALTRA-1 on CyberPvP, etc.) - plus the post-training where we ship new weights every week. $4/M is cheap in my opinion because I value things like privacy more than anything else: I don't want my customer engagement data, my prompts, my IP, to be logged, stored and used for training on Anthropic/OpenAI servers that's a no for me and for the serious enterprise customers I engage with. support the right teams, support the right companies. if $4/M is the only argument you have to not use CyberKimi, then I think CyberKimi is simply not the AI model for you or you don't get the idea. Go buy your DGX cards, install deepseek and keep burning your tokens. please contact me when you want to upgrade to something more serious like an exponentially capable cyber model like CyberKimi
4
304
Had one of my previous CVE's get marked as dupe (From 2024 I think?) to internal findings too
Weird things going on with ASB. One of my bugs has been said as addressed and acknowledged without CVE but the patch is still pretty much unreliable. Then some of my bugs have been marked as duplicate to their internal findings. Even my 2-year-old bugs have been marked as potential duplicate.
1
14
1,549
Wasn't this also done back in Mar/Apr of this year? (Not by XBOW) - IIRC had to do w/ Angle or a GPU bug that was a regression?
This just in: XBOW's Native Team has claimed the first Chrome Full Chain Exploit Bonus this year. Only four will be awarded in 2026. ⚡️
1
4
1,315
Capped: At ~$1,800 spent - bug remotely reads, but need hardware to actually get injection working remotely. Maybe if I do more radio work I'll shell out for stuff. A lot of progress made (60% autonomously? Roughly estimated). More experiments to come :)!
Update: at ~$1,154 spent - all bugs but 1 chain parked(not exploitable), BUT this one is a remote bug that's repro'd as a read + write (write is less deterministic so far unfortunately). Now moving to real hardware to see how well the agent does exploiting it over the network.
2
272
Update: at ~$1,154 spent - all bugs but 1 chain parked(not exploitable), BUT this one is a remote bug that's repro'd as a read + write (write is less deterministic so far unfortunately). Now moving to real hardware to see how well the agent does exploiting it over the network.
My agent just spent $543 overnight to come up with 4 leads (Kernel bugs in *OS) from the ~35 it claimed it found. Lots of room for improvement here but not as bad as I was expecting.
1
4
734
My agent just spent $543 overnight to come up with 4 leads (Kernel bugs in *OS) from the ~35 it claimed it found. Lots of room for improvement here but not as bad as I was expecting.
1
624
In ~2weeks from doing no research to 2 kern r/w bugs and a handful of bugs that are meh but should probably be fixed. (macOS only, wanna get the SRD again ;D). But the amount of handholding is nuts! But getting POCs put together really fast is super nice
1
3
422
My biggest skill in bug hunting/life in general has been my willingness to keep trying. @chompie1337 has an old writeup quote I love to tell people: you need to keep looking.
Same with bounties… lots of researchers completely omit the fact that some of the best bugs are found on sites from good recon or *excellent content discovery*, which are the unsexy parts of bounties. There’s also a bunch of failure omitted when you read the final write ups or listen to the interview in a podcast. Keep trying, keep failing, keep going.
2
14
108
14,600
If you build or maintain fuzzers for a living: what's the part you'd happily never deal with again? Harness rot, fleet babysitting, triaging noise, something else? Not selling anything, just trying to figure out if the boring grind is a real enough problem to solve.
1
144
Thinking effort doesn't fix hallucination. Even the best frontier model at matched HIGH still gets 24.2% of fields wrong on adversarial insurance docs. Going from default to HIGH buys 0-2pp per model. aginor.ai/extraction-test/
139
GPT-5.5 reported $405.9M of revenue on a document that says $95M. GPT-5.4 said $40.6M on the same page. I built 148 adversarial insurance documents to test five frontier models. The numbers got weird.
1
1
165
On visually-degraded pages, both OpenAI models fabricate numeric values at 2.6-6.5x the rate of Opus 4.7 and Sonnet 4.6 at matched default effort. Numeric hallucination rates: GPT-5.5 3.7% GPT-5.4 3.9% Sonnet 1.4% Gemini 0.8% Opus 0.6%
1
80
Codex (5.3xhigh) is a lot closer to CC than when I first used it; hope the gap continues to close.
1
178
Building a synthetic insurance doc generator for AI training. Need commercial property underwriters to look at sample loss runs and tell me what looks fake. 20 min, paid. DM me.
159
The SRD workflow is 1000x better than the traditional Apple developer workflow; and it pains me that they haven't replaced their "normal" workflow yet
i have to login to Sentry's Apple developer account multiple times a year, literally "I", because somehow I am the only person who can accept the continuous terms of services updates somehow the ios development world is stuck in the 2000s
1
343
We just shipped a big Oktalog update. You can now generate Okta auth logs that are already correlated with Jira issues and GitHub commits/PRs. One ZIP. Time-aligned events. Built for people who actually have to investigate incidents, not just look at dashboards.
1
1
182
Use it to: • Test SIEM rules without lab setup • Train SOC analysts on realistic scenarios • Train anomaly models on correlated auth + activity data • Demo security tools without touching customer data Try it: oktalog.getkashikoi.com
1
134
The advice we've gotten from Racoons has been very useful - would recommend. And it's free :)
In January 2025, I quit my job. Not for another offer. Not for a sabbatical. To build something I couldn't stop thinking about. Today, after a year in stealth, I'm launching @RacoonsAI Here's why: 🧵
1
1
227
Relaunched Oktalog, our free Okta log generator for security & eng teams. Generate realistic auth sequences, 20 logs no signup / 200 with a free account. Next up: chained, cross-app events for richer training data. Try it: oktalog.getkashikoi.com
109
We took the simulation approach that controls fusion reactors and applied it to security. Our Okta Log Generator creates realistic auth sequences: login→MFA→activity→logout with consistent context across events. Not random synthetic data. World models built from real security expertise. Security devs: Test your SOC rules properly. Enterprise: This verification approach scales to any domain. Try free: oktalog.getkashikoi.com
1
1
163