I'm poor now. 😆
Alexis Brignoni :python: :donor:
Digital forensics nerd. Opinions are mine and subject to change. He/Him.
#DFIR #DigitalForensics #FOSS #Python #OSDF #MobileForensics
Whoami?
🖥 Developer of open-source tools for digital forensics (iLEAPP, ALEAPP, & RLEAPP.)
📱 Digital forensics examiner specializing in mobile forensics.
🔊 Public speaker on cybersecurity and digital investigations.
🃏 Certified Digital Forensics Memer
Uppercase and uppercase +
Some stats from our Digital Forensics Now Podcast audio version for 2025.
Thank you so much for geeking out without this past year. Let's keep it going on 2026!
https://digitalforensicsnow.buzzsprout.com/
Coming back to work after the holidays.
Recently I learned that media generated by known #AI services will add provenance information in IPTC and C2PA metadata fields. In order to learn about these fields I made a script that leverages exiftool.org and C2PA python libraries to highlight these AI fields easily. Use the --json and/or --flattened arguments to show all fields.
The script can process one file or all files in a directory. This is an easy way to detect AI manipulation in unsophisticated scenarios.
The code can be downloaded here:
🎄My family knows me very well. Computers + Legos = Happy Me
🎁#HappyHolidays & #MerryChristmas to all. May 2026 be all you hope for.
Sometimes you have to align it just right...
I got the The Oracle archetype!
https://infosec.exchange/@abrignoni/wrapstodon/2025/3839f1aa0cb0d5d6
Necessity is the mother of invention. In my case it was the reason to put a python wrapper around the ffprobe functionality from FFMPEG.
The script gives frame counts, frames per second, extracts frame pixel content with hashes, and much more.
Interesting opinion from the Eleventh Circuit regarding video frames and sentencing guidelines in CSAM cases:
https://media.ca11.uscourts.gov/opinions/pub/files/202310697.pdf
Get the script here:
https://github.com/abrignoni/frame-counts-galore
What I hear vs what I see.
Same with how many drives are imaged and/or how many devices are extracted.
"According to current neuroscience, human thinking is largely independent of human language — and we have little reason to believe ever more sophisticated modeling of language will create a form of intelligence that meets or surpasses our own."
"One of generative AI's biggest risk factors, the ability to inspire confidence independent of reality."
Read The Register's Article here: https://www.theregister.com/2025/11/24/opinion_column_vibe_coding/
Validate your tools & processes, verify your data inputs. It is a must on #DigitalForensics & #MobileForensics.
Check out Six Steps To Successful Mobile Validation for more information on the topic ➡️ https://www.sans.org/white-papers/six-steps-to-successful-mobile-validation
Alexis Brignoni :python: :donor: boosted:
AAAAAAAAAAAAAAAAAAAHHHHHHHHHHHHHH
@w00p Your example is totally on point. 🎯
In #DigitalForensics:
📍Not all data sources are the same quality.
📍Looking like a coordinate is not the same as being one.
📍A found coordinate does not mean that the phone was there.
📍If you don't test you assume. And when you assume 🫏
But many claim we should use LLMs in #DigitalForensics, where reliability and repeatability are...
But they did bring me the SDCard adapter with no SDCard in it.
#DigitalForensics #DFIR #MobileForensics #AnySimilarityToTrueEventsIsCoincidental
🎃 Happy #DigitalForensics Halloween 👻
Client Info
Server: https://mastodon.social
Version: 2025.07
Repository: https://github.com/cyevgeniy/lmst