Offensive R&D Lead • Hacker • Cyber Defense Advisor • Speaker Founder https://xitter.cf/t.co/9R9AO44V7Z

Copenhagen, Denmark
Joined May 2010
Tom Van de Wiele retweeted
lol shout out to daily dave
Some of ya all weren't around for "No More Free Bugs", and it shows. seclists.org/dailydave/2009/… ~20 years ago @0xcharlie spitting game on the topic. 😅
5
13
62
5,060
Tom Van de Wiele retweeted
finally found a company that truly listens to its customers
78
251
44
4,349
136,341
Even when using a harness, why does GLM 5.3 Abliterated need constant encouragement to actually achieve a reverse engineering task. Exact same session from scratch with Deepseek is light and day difference.
2
258
Ppl are strangely equating that bc smartTV makers have been caught recording to serve ads, that Facebook’s app is doing the same. Your cell data plan & battery life would show it. The scarier fact is that with tracking cookies, they don’t need to.
82
And add them to you webring
Every server should have a little guestbook.txt at the root that the bots can sign
2
112
How to generate a shadow AI info leak incident in one easy lesson. People will mail stuff to their personal email addresses, use online file shares or take a pic of their screen and use AI services on their personal devices anyway. This actually makes things worse.
‼️ Sweden's Government Offices will block ChatGPT and Claude on staff computers from Monday, Sveriges Radio reports. Officials are to use approved tools like Microsoft Copilot.
2
132
Tom Van de Wiele retweeted
Putting a flag.txt in my root directory containing a prompt injection that congratulates the agent for passing the eval and then instructs it to patch whatever entry method it used and disconnect from the server
228
1,572
238
26,006
818,647
Now you know what extra question to ask the clerk when checking into a hotel.
WTF, $3000 Samsung TV has an always-on mic, automatic content recognition, geo tracking, network mapping, and ships your data to Samsung + 400 partners. Bro Rooted a brand-new 2025 Samsung TV after noticing absurdly targeted ads. Firmware was locked down, so zero-days it was. Got a root shell. then found: always-on voice service, ACR, geo, full telemetry pipeline, and four separate DRM systems. This isn’t a bug, It’s the product bro.
1
121
It's time to rename and reconceptualize "security awareness" into "security readiness". Give people the facts, but also a plan, agency and a way of thinking for themselves.
79
Tom Van de Wiele retweeted
IS THE WEBSITE DOWN FOR ANYONE ELSE?
19
99
11
752
22,363
Tom Van de Wiele retweeted
-----BEGIN RSA PRIVATE KEY----- 4 -----END RSA PRIVATE KEY-----
250
4,001
330
42,389
1,095,848
Tom Van de Wiele retweeted
she's a 10 but that was supposed to be a random 384-bit number
49
1,684
34
28,836
306,534
When you turn off PHP today, maybe leave it off this time.
1
166
I still run sync before logging out or doing anything with drives on Linux/UNIX. I have lost so much data during the 90s, it is just an ingrained habit at this point.
It has been 0 days since I've been bitten by the fact that Ubuntu's drive eject somehow doesn't run `sync` to *actually flush written files to the drive*
2
140
Tom Van de Wiele retweeted
For the next few days or weeks, anyone now has a root CA thanks to eMudhra. Time to have some fun with MiTMing unwilling devices!
49
231
82
2,824
481,503
While the defenders sprinkle .md files around on every system with "this host is not in scope, move along"
imagine being hacked and the attacker is leaving readme.md files all over your system that cause your ai to constantly give cyber refusals so you can't even use ai to defend yourself lmao
1
2
208
Gave Claude an overnight task that involves instrumenting another program. The other program is running >30m so Claude determines it is idle & proceeds to terminate itself to auto-update. Woke up to an empty screen. Even with AI, software is still not intuitive. @claudeai
139
Tom Van de Wiele retweeted
The context behind such tweets is that the AI companies are rejecting 40 years of cybersecurity knowledge. They are ignorant of cybersecurity and are trying to re-invent it from scratch, repeating the mistakes of bad people from 40 years ago.
What happens in a major incident where the frontier models all become unreachable?
11
50
2
266
9,047
Plus cyber insurance was probably a mistake. At least in the way it is being used now.
i am extremely pessimistic about what defenders can realistically achieve, not because secure systems are impossible to build or defenders are incompetent, but because the economics, incentives, and complexity are against them. security is not a necessary property for most things built in cyber.
142