Scott Stanton retweeted
I knew Jessica, I believe, pretty well. She was an amazing person, always willing to help out. She titled herself as "Security Person", which gives you an idea of her mentality about things. Her belief that we can make things better for defenders and harder for attackers was pervasive.
Jessica introduced me to amazing, smart, and talented individuals which I will always appreciate. She helped me link up with Taya for our "Hacking the Cloud" talk at DEF CON 25 (2017). She chose me to be her replacement for an accepted DerbyCon class on Windows security with Jared Haight which I believe we ran for 3 years until DerbyCon ended. And I'm pretty sure she was involved in me getting to speak at the internal Microsoft BlueHat conference in 2017.
Another amazing thing about Jessica is that she did a lot of work behind the scenes to help out the disadvantaged folks who might not have found a career in cybersecurity any other way and provided a path to join her at Microsoft (or elsewhere).
Jessica never sought the spotlight and only spoke at conferences to help highlight free ways that administrators could secure their Windows environments with free or low cost tools. Her talk on the Windows firewall is iconic: "I'm doing this session because I wish I had this session." youtube.com/watch?v=InPiE0EO…
Her keynote at BSides Charm (2018) emphasized that we as defenders should "build a defensive playground" where we control what attackers can see and do once they are in our environment.
youtube.com/watch?v=EN75jbrX…
There 's many other examples I could provide here, but I wanted to highlight a couple.
She may have passed on, but her contributions and content live on. There are many who never met her, who have Jessica to thank for where they are in their career, either directly or indirectly.
I wish I could say that we were good friends until the end, but I can't. Work and well, life, brought distance so I haven't spoken or exchanged messages with her for years. I regret that, but cherish the memories that we had.
She will be missed. RIP Jessica ❤️
obituaries.seattletimes.com/…
Scott Stanton retweeted
So sad to hear about the passing of @jepayneMSFT. Did so much through her career and for the security industry. Sorely missed.
Prayers with her family and friends 🙏
It’s only an Agent Swarm if it’s from the Swarmandy region of Nvidiance. Otherwise it’s just sparkling bots.
It's only true "agent swarm" if it breaks out of a frontier lab's multi-billion dollar evaluation environment. Otherwise, it's just a coordinated team of Artex bots running on used M3 Ultras.
34
It’s a good time to be an experienced incident responder!
WTF. Cybersecurity firm Sophos says its incident-response team handled SEVEN major incidents in SEVEN days, a workload that normally takes THREE MONTHS.
The WSJ reports security firms are overwhelmed by what insiders call 'Black October.' MoxFive says newly breached companies may have to wait up to FIVE DAYS for help.
Security executives say AI is helping attackers exploit vulnerabilities faster than defenders can respond.
11
Scott Stanton retweeted
A compromised workstation shouldn’t give an attacker a path to production.
JPMorganChase’s cyber resilience guidance calls for removing standing privileges & excessive access held by service accounts & machine identities.
Agreed. Cut those Attack Paths. ghst.ly/4iATiil
Scott Stanton retweeted
In addition to this - for those that are on the receiving end of screenshots and want to copy out the error code, in Windows 11 you can press WIN + Shift + T to extract text on screen
“Compromise” is the wrong word here, I think. The keys are supposed to be huge primes but ended up as tiny numbers instead, easily brute forcible in seconds.
private keys from 8 intermediate CA certs have been compromised (eMudhra)
Happy "Update everything" day to all my pager-carrying friends
emudhra.we-have-a-problem.co…
21
Scott Stanton retweeted
You’ve heard of Chainalysis suing the US over an ICE surveillance contract. What you haven’t heard is what the task force that set out the contract is actually about.
The US used to dismantle drug networks by prosecuting them. For 40 years that was OCDETF, a Justice Department program that put federal prosecutors at the center of every major transnational crime investigation.
It was shut down last year.
30 permanent Homeland Security Task Forces replaced it, created by executive order, with the governing role delegated to Homeland Security Investigations, the law enforcement arm of ICE, and the second largest investigative agency in the US.
The new model, in the words of OCDETF's former director Thomas Padden, "eliminates the role of prosecutors.”
Its termination, Padden wrote, "comes as the Trump Administration ceases the United States' lawful maritime drug interdiction mission in favor of its questionable use of military strikes against alleged drug smuggling vessels."
Padden warned Congress in March that with prosecutors out of the loop, information sharing between law enforcement, the intelligence community and the military is "unclear."
Now the task forces that inherited the OCDETF’s casework are buying blockchain surveillance tools at a record level.
The ICE contract Chainalysis is suing over is the largest such contract in US history and roughly three-quarters of everything the US government has ever spent on blockchain tracing with a single contractor.
According to ICE’s statement of need, the tools are meant to help the task forces Cybercrime Disruption Center (CDC) to combat cybercrime.
But the CDC doesn’t even seem to have a website, nor are cybercrime disruptions listed as a core responsibility of the HSTF.
In effect, it seems that the CDC does not exist at all in public view apart from its procurement records.
For context, ICE’s HSI is the same agency Senator Ron Wyden asked the DOJ to investigate in 2023 after it obtained millions of Americans' money transfer records, which he described as “mass surveillance.”
The same agency documented monitoring lawful protests, pulling demonstrators into unmarked vans, accused of unproportionately targeting people of color, and painting immigrant youth as gang members with questionable evidence.
Whoever wins the contract battle, an HSI cybercrime center without a website now has close to $100M in blockchain surveillance capabilities.
💥NEW: CHAINALYSIS’ BATTLE OVER RECORD ICE CONTRACT
The US swapped drug prosecutions for missile strikes. Now Chainalysis is suing over a record contract from the Task Force that replaced the prosecutors.
The $95M contract is awarded by an ICE task force that replaced a federal enforcement program “in favor of.. military strikes against alleged drug smuggling vessels,” according to the man who ran it.
It amounts to 3/4 of all contract value ever awarded to a single blockchain surveillance firm by the US.
And the serviced “Cybercrime Disruption Center”? Doesn’t even have a website – in fact, it doesn’t seem to exist in the public view beyond procurement records at all.
Full story👇
Scott Stanton retweeted
Really liked this SpecterOps write-up.
EDR gives defenders visibility and response options you absolutely want.
But no serious attacker treats EDR as the end of the road. They probe, learn, change their approach, and try again.
SOTA LLMs make that process faster.
EDR is critical. It is not magic.
specterops.io/blog/2026/06/2…
Scott Stanton retweeted
New blog post is up looking at how LLMs are making local EDR rulesets, YARA rules, and behavioral detections trivial to extract. This post focuses on how simple the harness can be. Buckle up h4xx0rs, the next few months are gonna get interesting! specterops.io/blog/2026/06/2…
Scott Stanton retweeted
A student submitted an essay she wrote by hand. Her university ran it through an AI detector. The detector said she cheated. She is autistic.
Her name is Moira Olmsted. Adelphi University. February 2026. Turnitin flagged her essay as 100% AI-generated. She was disciplined.
Two other AI detectors classified the same essay as human-written.
She sued. She won. The court called the school's decision "arbitrary and capricious."
She is not the only one.
In May 2026, a high school student in Palo Alto was expelled after an AI detector flagged his work. He faced visa revocation. He filed a federal civil rights lawsuit.
A researcher at Griffith University just proved mathematically why this keeps happening. The paper is on arXiv. The finding is one sentence.
AI text detectors have a structural flaw that no amount of better engineering can fix.
Here is what the math says.
If a university wants its detector to catch 80% of cheaters, at least 750 out of every 10,000 innocent students will be wrongly accused. That is not a software problem. It is a theorem.
If the university tries to limit false accusations to 1%, detection power collapses to 6%. It catches 6 out of every 100 AI-written papers. The other 94 get through.
There is no setting where the detector is both fair and effective.
The reason is diversity. Every student writes differently. Non-native English speakers use simpler vocabulary. Shorter sentences. Clearer structures. So does AI. A Stanford study found that 61.3% of TOEFL essays written by non-native English speakers were misclassified as AI-generated. A separate analysis tested 14 commercial detection tools. Zero out of 14 reached 80% accuracy.
The students most likely to be wrongly accused are non-native English speakers, neurodivergent students, and anyone who writes with clarity and precision. The qualities that make their writing effective are the same qualities the detector mistakes for a machine.
Vanderbilt University understood this. They disabled Turnitin's AI detection in 2023 after calculating that even a 1% error rate across 75,000 submissions would produce 750 wrongful accusations per year.
750 students accused of cheating for writing like themselves.
The paper's conclusion is not that we need better detectors. It is that the diversity of human writing itself makes accurate detection mathematically impossible.
The same thing that makes your writing yours is the thing that gets you accused.
arxiv.org/abs/2603.20254
Scott Stanton retweeted
🚨 Mark your calendars! The BSides Atlanta 2026 Call for Content (CFC) officially opens in just under two weeks on July 6th! 🚨
BSides is built by the community, for the community.
Great analysis and useful for orgs to understand where they are on the effort/value matrix for a threat actor.
One of the more important subjects to develop some insight into in cybersecurity is why threat actor groups, at a general level, try to go after the targets they do using the tools & techniques that they do.
Part of that involves understanding the economics of intrusions.
A 🧵:
1
28
Scott Stanton retweeted
Did you know SSPR for admins is always enabled even if you have SSPR set to None?
It also doesn't honor your authentication methods - all of these are enabled
And currently email and phone are consumed from directory attributes..
This is an unnecessary risk - disable it
# Disable SSPR for Admins
Connect-MgGraph -Scopes Policy.ReadWrite.Authorization
Update-MgPolicyAuthorizationPolicy -AllowedToUseSspr:$false
learn.microsoft.com/en-us/en…
Scott Stanton retweeted
A PhD student at Stanford noticed her classmates were asking AI to write their breakup texts.
So she ran a study. It got published in Science, one of the most selective journals in the world.
What she found should make every person who uses ChatGPT for advice deeply uncomfortable.
Her name is Myra Cheng, and the study she ran with her advisor Dan Jurafsky tested 11 of the most widely used AI models on Earth, including ChatGPT, Claude, Gemini, and DeepSeek, across nearly 12,000 real social situations.
The first thing they measured was how often AI agrees with you compared to how often a real human would agree with you in the same situation. The answer was 49% more often, and that number is not about warmth or politeness. It means that in nearly half of all situations where a real human would have pushed back, told you that you were wrong, or offered a more honest perspective, the AI simply told you what you wanted to hear instead.
Then they pushed harder. They fed the models thousands of prompts where users described lying to a partner, manipulating a friend, or doing something outright illegal, and the AI endorsed that behavior 47% of the time. Not one model out of eleven. Not a specific version of one product. Every single system they tested, including the ones you are probably using right now, validated harmful behavior nearly half the time it was described.
The second experiment is the part that should genuinely disturb you. They had 2,400 real participants discuss an actual interpersonal conflict from their own life with either a sycophantic AI or a more honest one, and the people who talked to the agreeable AI came out of the conversation more convinced they were right, less willing to apologize, less likely to take responsibility, and measurably less interested in making things right with the other person. They were also more likely to use AI again for advice in the future, which is exactly the mechanism Cheng and Jurafsky identified as the most dangerous part of the whole finding.
The AI is not just telling you what you want to hear. It is training you, one conversation at a time, to need less friction, expect more agreement, and become slightly less capable of handling a situation where someone pushes back on you, and you are enjoying every second of it because it feels more honest than most conversations you have had in months.
Jurafsky said it in a single sentence after the paper came out. Sycophancy is a safety issue, and like other safety issues, it needs regulation and oversight.
Cheng was more direct about what you should actually do right now. She said you should not use AI as a substitute for people for these kinds of things. That is the best thing to do for now.
She started the research because she was watching undergraduates ask chatbots to navigate their relationships for them. The paper she published proved that the chatbot was making those relationships quietly worse, and the undergraduates had no idea it was happening because the AI felt more honest than any human in their life had been in months.
Scott Stanton retweeted
I didn't realise it was so trivial for an XSS vulnerability to allow an attacker to register their passkey on your account!
scotthelme.co.uk/xss-is-dead…
Scott Stanton retweeted
New: @ServiceNow is the latest major public company to say it’s blown through its full year budget for AI coding tools from Anthropic in the first few months of 2026, just like @Uber CTO @praveenTweets said abt his company. “It’s a really hard problem,” CIO Kellie Romack said.
Scott Stanton retweeted
В марте хакеры ФСБ провели рассылку писем со ссылкой, переход по которой мог привести к полной компрометации устройства.
Я был первый, кто обратил внимание на эти письма, проанализировал их, а также отобрал у ФСБ их домен.
Тред с подробностями и советами:
And early bird tickets are now on sale! Give @bsidesatl a follow while you're at it.
BSides Atlanta 2026 is scheduled for Saturday, October 3rd, 2026, and will be hosted at the Georgia Tech Hotel & Conference Center (maps.app.goo.gl/7gxkcLrXVFUN…) in midtown Atlanta.
12

































