Jake Williams

Jake Williams

@malwarejake.bsky.social

Breaker of software, responder of incidents, IANS Faculty, VP R&D Hunter Strategy.

15.1K followers520 following9.9K posts

Jake Williams
@malwarejake.bsky.social

AI leaders: if you want to be taken seriously, come with actionable solutions, not "you'll have to use AI to solve this problem we created, but we don't really have a plan for HOW AI will solve this."

breachplease-fm.bsky.social@breachplease-fm.bsky.social

Sam Altman: we can't 'patch our way out of this.' Jake and Jess on AI leaders lecturing defenders — plus FDA's accountability fight over Gen AI medical devices and why non-determinism doesn't belong near a pacemaker. Watch: youtu.be/zapjfPeHCLQ @malwarejake.bsky.social @secitup.bsky.social

Jake Williams
@malwarejake.bsky.social

Late night, I saw Unit 42 retracted their claim of post-exploitation activity, but now several others are confirming it. Turns out, nothing we recommended was unique to Unit 42, so 🤷‍♂️

breachplease-fm.bsky.social@breachplease-fm.bsky.social

Unit 42 confirmed it: PHP exploitation in the wild, domain admin in hours, millions of services exposed. Jake and Jess cover the threat-hunting framework, the 4 ways to treat risk, and why AI is speeding up tech debt. Watch: youtu.be/OTu_vpLaPW8 @malwarejake.bsky.social @secitup.bsky.social

Jake Williams reposted

Bradley P. Moss
@bradmossesq.bsky.social

They have nothing.

Aaron Rupar@atrupar.com

BASH: What's the crime? AG TODD BLANCHE: What's the crime? Well it depends. There *could* be a crime. But the real wrong here is that Alaskans are gonna be confused. BASH: The Alaskan Supreme Court ruled 4 months ago on this matter

Jake Williams reposted

Phil Hagen
@philhagen.com

In three weeks: “Teams disconnects 72% of all legitimate calls using flawed AI. Teams users rejoice.”

InfoSec@infosec.skyfleet.blue

Microsoft Teams to Use AI for Detecting Phishing Attacks Using Teams Conversations

cybersecuritynews.comMicrosoft Teams to Use AI for Detecting Phishing Attacks Using Teams ConversationsMicrosoft is preparing to add AI phishing detection to Microsoft Teams, using conversation context to spot scams that may contain no harmful links or attachments. The feature will focus mainly on first conversations with external contacts, where attackers can pose as trusted support staff and ask employees to share sensitive information. Microsoft detailed the change in Message Center notice MC1519866 , published on October 21, 2026. Targeted Release is scheduled to begin in early November and finish by mid-November. Worldwide general availability is expected to start in early December and finish by mid-December, rather than reaching every customer in November. How Conversation Context Helps Detect Phishing Traditional warning signs, such as suspicious links and infected files, are not always present in social engineering attacks. A message can instead build trust, claim an urgent account problem, and persuade someone to disclose a password or approve remote access. Microsoft says the new capability will analyze conversation context to identify phishing, impersonation, social engineering, and deceptive intent. Its support guidance describes coverage for individual and group chats involving new external contacts across mobile, desktop, and web. Roadmap entry 575288 lists Android, iOS, Mac, desktop, and web support for the worldwide standard cloud. The warning example accompanying the announcement shows an external account called “MOD Administrator” discussing permission problems before asking, “Can you share your credential first?” Teams displays a notice saying the message could be a phishing attempt. The example illustrates why the meaning of a conversation can matter even without a dangerous URL. Source : Microsoft This addresses a familiar attack pattern. Cybersecurity News previously reported on attackers impersonating IT help desk staff through Teams , using convincing support requests to push users toward malware or remote access. Separate reporting described fake support operators abusing Quick Assist after contacting employees through external Teams accounts. Microsoft says detection may initially run silently while it checks quality. Visible warnings may appear after that validation, so users should not expect every suspicious message to trigger an immediate banner during the rollout. For most users, protection centers on external first-contact chats. Teams mobile users in China may also receive protection for conversations within their own organization, reflecting fraud patterns seen in that region. This is a specific extension, not a promise of universal internal chat scanning. Microsoft states that the capability serves security purposes only. It is not intended for advertising, employee monitoring, productivity analysis, or conversation summaries. Customer message content is not retained by the detection system for model training or other secondary uses. That statement should not be read as a change to all existing Teams retention policies. What Organizations Should Do No new administrator configuration is required to enable the protection. However, Microsoft recommends reviewing guidance for unfamiliar external contacts and encouraging users to report suspicious activity or incorrect detections. When a warning appears, employees should check the sender’s identity and email address through a trusted channel. Microsoft advises blocking and reporting confirmed phishing, while mistaken warnings can be reported as not a concern. Users should never share passwords, verification codes, financial details, or sensitive information through an unexpected support conversation. The update adds another warning layer, but staff still need to independently verify unexpected support requests and report suspicious contacts, rather than treating the absence of a banner as proof of safety. Update defenses against new malware as soon as it appears. Integrate TI Feeds with data from 16K SOCs . The post Microsoft Teams to Use AI for Detecting Phishing Attacks Using Teams Conversations appeared first on Cyber Security News .
Jake Williams
@malwarejake.bsky.social

By definition, rubber stamping isn't leadership. It's the opposite of that.

Aaron Rupar@atrupar.com

Bernie Sanders: "If you believe as I do that Donald Trump is the most dangerous president in the history of the United States, we need to get rid of that rubber stamp Republican leadership in the House and Senate."

Jake Williams
@malwarejake.bsky.social

Nationalize SpaceX NOW.

Dare Obasanjo@carnage4life.bsky.social

Ukraine’s military Starlink service is now working after a 10-day disruption that affected its drones. Musk claims Ukraine violated Starlink’s terms of service by using it for offensive drone strikes. Trump reportedly called Musk to ask him to restore it. SpaceX can now change the course of wars.

Jake Williams reposted

Jake
@jake.sooscreek.productions

Mythos attempted nothing. Someone, wrote an MCP harness that instructed Mythos to do these things

This post is unavailable.

Jake Williams reposted

Phil Hagen
@philhagen.com

Every single person that accepts a pardon from this criminal administration must never know a moment of peace for the rest of their lives. They must be globally shunned and ridiculed without reprieve.

Ana Marie Cox@anamariecox.com

I guess this is one way to keep young people interested in working for Republicans. "So, yeah, benefits include 401k, FSA, a full pardon, and dental."

Jake Williams
@malwarejake.bsky.social

Even if you look at the still and your brain says "it is obvious what you'll see if you click on this, skip it" - trust me, you owe it to yourself to do it anyway.

This post is unavailable.

Jake Williams reposted

Ian Coldwater 🧊🚫
@lookitup.baby

Update: I’m now the first person to have said both “chinga la migra” and “stay safe, fuck ICE” on stage at a tech conference

Aditya Mukerjee 🦦 🏳️‍🌈@chimeracoder.bsky.social

Congratulations to @lookitup.baby for being the first person I've heard say "chinga la migra" on stage at a tech conference 🔥🔥🔥 @blueteamcon.com

Jake Williams
@malwarejake.bsky.social

Dear Cuban citizens, Assurances were made to Iranian people who sought to "create positive change" less than a year ago. I'd tell you to ask them whether the US held up its end of the deal, but they're in prison or dead - mostly the latter.

Jake Williams
@malwarejake.bsky.social

THANK YOU.

Tanya Janca | SheHacksPurple@shehackspurple.bsky.social

"Developers should carefully investigate every dependency before they install or update it." Okay - Sure... And while we're at it, perhaps they could personally audit the source code of all 1,847 transitive dependencies before lunch. 😂 We need security advice that works in the real world. 1/5

Jake Williams reposted

Morgan J Freeman
@mjfree.bsky.social

Well, well, well…. An analysis of voter registration records reveals Republican operative Scott Presler had 3 out-of-state voters use his Pennsylvania address to vote in the 2024 election. This is what voter fraud looks like.

Jake Williams reposted

Ed Bott
@edbott.com

The replies are great.

Reuters Pictures@pictures.reuters.com

Lights are reflected off a bulletproof screen as President Trump speaks during a campaign rally in Greenville, North Carolina. More photos of the week: reut.rs/4yGkBNh 📷 Evan Vucci

Jake Williams
@malwarejake.bsky.social

Since this is getting a lot of comments, I'll give you one more hot take: I believe it's an abject failure of cybersecurity programs that we consider users so crucial to phishing defense that we feel the need to regularly phish them at all - regardless of the pretext used.

Jake Williams@malwarejake.bsky.social

Fire everyone on the cybersecurity team who had anything to do with this. What are you even testing here?

Jake Williams
@malwarejake.bsky.social

I hate that people sometimes say my writing sounds like AI and I have to explain that AI companies stole my writing and that of so many other professionals. We don't sound like AI, AI sounds like us.

Mike Masnick@masnick.com

Just wrote an article, and as I'm doing an edit, I spot that I've written a "not x, but y" kind of sentence, and immediately decide I need to change because that's become so identified as AI speak. *sigh* The sentence sounded better that way, though.

Jake Williams
@malwarejake.bsky.social

"May" is doing a LOT of heavy lifting in this headline.

Gillian Brockell@gillianbrockell.com

HOLY SHIT. In a sworn statement yesterday, an ICE agent told a federal judge that an ICE flight to Equatorial Guinea had been cancelled and so couldn't be used to bring Ahmed Soliman back to the United States. The plane just landed in Equatorial Guinea. prospect.org/2026/10/23/t...

prospect.orgThe Trump Administration May Have Just Lied to Judge About an ICE flight to Equatorial Guinea - The American ProspectAn ICE agent told a judge a charter flight to Equatorial Guinea had been canceled. It just landed there.
Jake Williams
@malwarejake.bsky.social

Fire everyone on the cybersecurity team who had anything to do with this. What are you even testing here?

shauna@goldengateblond.bsky.social

this is fucking appalling. i already knew james dolan (who owns the knicks and madison square garden) was a douche but this is next-level. www.nytimes.com/athletic/771...

Jake Williams
@malwarejake.bsky.social

I'm beginning to think that he legitimately doesn't know what's happening in his administration.

Aaron Rupar@atrupar.com

Trump on Katie Zacharia: "She was a perfect, beautiful woman. When she walked in to the press room and they started screaming at her, she freaked out. And I said, 'You know what? There are nicer ways to make a living.'"

Jake Williams reposted

Thomas Brewster
@thomasbrewster.bsky.social

🚨NEW🚨 Over the last 2 months, OpenAI's hacker lab coordinated disclosure of *those* PHP bugs. I got an inside peak at that team's work, hearing about the 100s of other wild flaws OpenAI's models are finding daily. Including fun things like Chrome 0-click 0days. www.forbes.com/sites/thomas...

forbes.comAt OpenAI’s Cyber Lab, Hackers Race To Protect The Internet From AI AttacksUsing the latest GPT model, cyber researchers in OpenAI’s Project Daybreak community found one of the most critical internet bugs in recent memory. It’s part of an all-out sprint to secure the web bef...
Jake Williams
@malwarejake.bsky.social

This is BY FAR the spiciest episode we've ever recorded. We agree a lot and disagree a lot too. @secitup.bsky.social was on FIRE. The one thing we agree on 100% is that if Chainguard's actions aren't met with some significant blowback, coordinated disclosure as we know it is dead.

breachplease-fm.bsky.social@breachplease-fm.bsky.social

A ChainGuard CEO broke a coordinated PHP disclosure embargo — and the industry's asking if coordinated disclosure just died. Jake and Jess break down why, plus a board framework beyond 'more budget.' Watch: youtu.be/WHewKFcNRfk @malwarejake.bsky.social @secitup.bsky.social

Jake Williams reposted

Pavel
@spavel.bsky.social

AI has finally used up all the money

Christopher Mims@mims.bsky.social

Wall Street is bailing on the AI trade. “Signals from Meta that it could soon go on a cash-raising spree helped trigger another massive rotation trade on Thursday, with investors dumping stocks connected to the AI boom and piling into almost everything else“ www.wsj.com/finance/stoc...

wsj.comMeta’s Hunt for AI Capital Strikes a Raw Nerve on Wall StreetThe tech-heavy Nasdaq fell 2.6%, while Treasury yields posted its biggest drop since 2023.
Jake Williams
@malwarejake.bsky.social

I have a doctor appointment first thing Monday and it can't come too soon. My pain level is back through the roof. I worked a 14+ hour day today, but working is the only thing taking my mind off my nerve pain. I hope she does trigger point or at least toradol. I'm getting desperate for relief.