iiinova — where your digital self keeps evolving

VessOnSecurity @vessonsecurity.bsky.social · 01/10/2026

Sorry, that was way before my time - I was just a 9-years old at the time. Try asking him?

100

VessOnSecurity @vessonsecurity.bsky.social · 30/09/2026

Maybe Russia was just worried that Karpov would beat Kasparov at chess again.

000

VessOnSecurity @vessonsecurity.bsky.social · 28/09/2026

Not sure about that... Trying to fly with 19 suitcases - if it's not a crime, it definitely ought to be.

010

VessOnSecurity @vessonsecurity.bsky.social · 18/09/2026

This is only a first iteration. Eventually, there will be a link to a feedback page where people will be able to submit additions/corrections.

010

VessOnSecurity @vessonsecurity.bsky.social · 18/09/2026

Don't be afraid to make a mistake - e.g., "Fnacy Baer". Press Enter after entering the APT name in the field. You get a list of aliases of that APT, MITRE ATT&CK reference for it, list of companies using these aliases, etc., with clickable links to the web pages the information is based on.

110

VessOnSecurity @vessonsecurity.bsky.social · 18/09/2026

I made a thing. APT name cross-reference database you can query: stats.nlcv.bas.bg/d/apt-name-c... Enter the apt name in the "APT name / alias" field (upper left corner). Optionally select a country. Can select only country without an APT name, to see all the APTs from that country.

stats.nlcv.bas.bg

Grafana

130

VessOnSecurity @vessonsecurity.bsky.social · 14/09/2026

In Anthropic's Claude, this setting is definitely off by default. I know, because I've turned mine on; I *want* my conversations to be used to improve the model. It is unfortunate that humans have to read them, though - I would have preferred this to be done by automatic machine learning.

020

VessOnSecurity @vessonsecurity.bsky.social · 14/09/2026

But what if I *want* my content to be used to improve the model for everyone?

010

VessOnSecurity @vessonsecurity.bsky.social · 09/09/2026

Well, such a request is hardly surprising. Imagine if human soldiers routinely said "Sorry Dave, I can't do that" to their commanding officers...

001

VessOnSecurity @vessonsecurity.bsky.social · 06/09/2026

Maybe they were afraid of being deported.

000

VessOnSecurity @vessonsecurity.bsky.social · 01/09/2026

Health (or lack thereof) update: blog.bontchev.nlcv.bas.bg/en?articleId...

blog.bontchev.nlcv.bas.bg

My battle with cancer

Bye

120

VessOnSecurity @vessonsecurity.bsky.social · 27/08/2026

This totally sounds like something the EU would do.

010

VessOnSecurity @vessonsecurity.bsky.social · 25/08/2026

Was it the word "sophisticated" that shocked you?

000

VessOnSecurity @vessonsecurity.bsky.social · 17/08/2026

I have no problem if the watermark says just "this was written by Claude". But it's hidden - who knows what else it contains? Like, which account requested it, where it is based, etc. - data, which is later sold to advertisers or abused by law enforcement.

100

VessOnSecurity @vessonsecurity.bsky.social · 16/08/2026

Health (or lack thereof) update: blog.bontchev.nlcv.bas.bg/en?articleId...

blog.bontchev.nlcv.bas.bg

My battle with cancer

Bye

240

Reposted by VessOnSecurity

Evan Greer @evangreer.bsky.social · 16/08/2026

there is no such thing as "age verification." All "age verification" is actually "identity verification." You can coat it in glitter and "privacy-preserving" wrapping paper. But in the end, mandating age verification means mandating surveillance be built into software and hardware. Period.

105138705603

VessOnSecurity @vessonsecurity.bsky.social · 14/08/2026

"Pax"?? As in, "solitudinem faciunt, pacem appellant"?

100

VessOnSecurity @vessonsecurity.bsky.social · 04/08/2026

Too early. China is not ready. They think long-term. Give it another 5-7 years.

000

Reposted by VessOnSecurity

Electronic Frontier Foundation @eff.org · 04/08/2026

On Wednesday, the Senate Commerce Committee will vote on four bills that would expand age verification, increase online surveillance, and make it harder to access lawful speech online. Take one minute to tell your senators to vote NO. www.acteff.org/campaign/16...

acteff.org

Tell the Senate: Don't Turn the Internet Into an ID Checkpoint

Tell the Senate to Reject Bills That Expand Age Verification

7215159

Reposted by VessOnSecurity

Asa Dotzler @asadotzler.com · 22/07/2026

A company that uses distillation to make its AI models should not be called "Moonshot," but rather "Moonshine." Ba-dum-tss

062

VessOnSecurity @vessonsecurity.bsky.social · 20/07/2026

Health (or lack thereof) update: blog.bontchev.nlcv.bas.bg/en?articleId...

blog.bontchev.nlcv.bas.bg

My battle with cancer

Bye

021

VessOnSecurity @vessonsecurity.bsky.social · 08/07/2026

Health (or lack thereof) update: blog.bontchev.nlcv.bas.bg/en?articleId...

blog.bontchev.nlcv.bas.bg

My battle with cancer

Bye

020

VessOnSecurity @vessonsecurity.bsky.social · 01/07/2026

They already have. Claude Sonnet 4.6 (one of the relatively low-grade "safe" models) has been giving me warnings every day because I use it for "dangerous" purposes like fixing bugs in the Cowrie honeypot. ☹️

000

VessOnSecurity @vessonsecurity.bsky.social · 01/07/2026

Yeah, we Europeans are big on eco stuff.

000

VessOnSecurity @vessonsecurity.bsky.social · 01/07/2026

Obviously, it's to fight pirates, what.

000

VessOnSecurity @vessonsecurity.bsky.social · 25/06/2026

*Watto voice*: Cancer always wins.

010

VessOnSecurity @vessonsecurity.bsky.social · 22/06/2026

Friend? No. But, as somebody noted, it's a combination of a wise mentor, and unpaid intern, and a garbage can full of rabid raccoons.

030

Reposted by VessOnSecurity

Lorenzo Franceschi-Bicchierai @lorenzofb.bsky.social · 22/06/2026

NEW: I took a look at the history of using export controls to limit the proliferation of cyber capabilities. From the 1990s Crypto Wars to limit the spread of PGP, to Wassenaar to stop spyware flowing out of Europe, and now powerful AI models, I argue that this type of regulation does not work.

techcrunch.com

From PGP to Mythos: a brief history of export controls that didn't stop anyone | TechCrunch

For the last 30 years, stopping the flow of cybersecurity-related software has proven to be ineffective. It's unclear why it would work now with Anthropic’s cybersecurity model Mythos.

53725

VessOnSecurity @vessonsecurity.bsky.social · 22/06/2026

Health (or lack thereof) update: blog.bontchev.nlcv.bas.bg/en?articleId... (Sorry about the delay, the room where the server is, is undergoing renovations and the power is occasionally off.)

blog.bontchev.nlcv.bas.bg

My battle with cancer

Bye

030

VessOnSecurity @vessonsecurity.bsky.social · 22/06/2026

If I'm around, the Nando's won't last ten minutes.

000

VessOnSecurity @vessonsecurity.bsky.social · 22/06/2026

Ah, yes, the worm that never dies. Although, TBH, I haven't seen it hitting my SMB honeypot for a long time (like, 2-3 years). Now do WannaCry. (Chart from that SMB honeypot for the past 3 months.)