Pinned Tweet
For decades, defenders have been outgunned. And real people paid the price. We started @TENEXai to change that: elite human expertise fused with AI, purpose-built to fight cybercrime.
We partnered with @freethinkmedia and @bigthink to tell that story.
It's Time to Protect.
Eric Foster retweeted
I’m still not sure people realize that targeted web app exploitation wasn’t widespread before LLMs. Sure, companies with bug bounty programs saw plenty of it, but that wasn’t reflected in how most financially motivated attackers operated.
What you’re seeing now is some level of automation being applied to a bunch of web apps and internet facing infrastructure that attackers largely ignored before. Historically, attackers focused on volume, exploiting common vulnerabilities in frameworks and widely deployed applications rather than spending time finding bugs in individual web apps.
This space has always found a balance, and I expect it will again.
Eric Foster retweeted
侵入から15分。人間のCSIRTが対応を始める前に、攻撃が終わる時代が来た。
Google Threat Intelligence GroupとMandiantが、AI攻撃フレームワーク「ARTEX」の実戦投入を報告。
これ、かなり重要な転換点だと思う。
ARTEX、PentAGI、CAIRNなどのマルチエージェント型攻撃ツールが、実際の侵害事案で確認されている。
しかも、単なる脆弱性スキャンではない。
・Active Directoryの偵察と権限昇格
・Kerberoasting、DCSync
・AWS/Azureの権限探索
・横展開と永続化
・データ窃取の準備
これらをAIエージェントが自律的に実行する。
ある攻撃グループでは、最初に侵害されたアカウントからシステム深部への到達まで通常約4時間、最短15分。
さらに、23秒間に27イベント、エラーが出ればコードを書き直して再試行。
人間の攻撃者とは時間軸が違う。
興味深いのは、AIの攻撃が必ずしも洗練されていないこと。
大量のログを残す。エラーも出す。試行錯誤も雑。
それでも速さと物量で侵害に成功する。
Googleも指摘しているが、攻撃手法そのものが新しいわけではない。
盗まれた認証情報、過剰な権限、未修正の脆弱性。
昔からある弱点を、AIが機械速度で連鎖的に突いてくる。
CSIRTが考えるべきは、アラートを検知できるかだけではない。
検知した時点で、すでに何が奪われているのか。
人間によるトリアージ、承認、封じ込めという運用が、この速度に追いつくのか。
攻撃が自動化されたのに、防御だけ人間の判断待ちでは勝負にならない。
これからのSOC/CSIRTはMTTDだけでなく、封じ込めまでの時間と、侵害後に攻撃者が実行できる操作の範囲を見直す必要がある。
Googleの報告は、AI時代のインシデント対応を考えるうえで必読。
cloud.google.com/blog/topics…
Eric Foster retweeted
Wiz co-founder and CTO Ami Luttwak says AI can now carry out cyberattacks in 20 minutes that once took skilled attackers weeks or months.
In his view, companies need to continuously attack their own systems with the latest AI models to stay secure. For startups, that means building automated testing and patching into the development process from the beginning.
Ami joins a16z's Sarah Wang and Zane Lackey to discuss what changed during Black October, why patching is falling behind, the cost of continuous security testing, and why better security architecture matters more than simply buying more tools.
0:00 Intro
1:04 Welcome to Black October
3:19 How AI made every day a zero day
6:51 The PHP vulnerability: 70% of the internet
9:19 Why patching can't keep up
11:32 How Wiz uses AI to find vulnerabilities
16:59 The cost of continuous testing
19:23 Cloud security in the AI era
22:18 The red agent: attack yourself first
26:41 Agent swarms attacking your own systems
28:43 What founders must do before they ship
32:15 Where security budgets should go
36:10 Most of what security teams do isn't security
43:34 Scan for Good and public infrastructure
YouTube: youtu.be/PbD9pVXxsYM
@amiluttwak @sarahdingwang @zanelackey
Eric Foster retweeted
As a big 'monitor the situation' guy, there are currently way too many situations to monitor.
Eric Foster retweeted
Threat actors are weaponizing ARTEX, an open-source, large language model (LLM)-driven agentic platform to automate multiple phases of cyber intrusions at machine speed. Since at least August 2026, a variety of malicious cyber actors have used ARTEX to conduct campaigns targeting organizations globally across multiple sectors, including financial services, government, and technology.
ARTEX significantly reduces the time required to chain attack techniques and enables threat actors with minimal technical expertise to conduct complex, multi-target operations simultaneously.
Today, the FBI released a FLASH highlighting the tactics, techniques, and procedures and indicators of compromise associated with the ARTEX framework. Organizations should follow the mitigation steps outlined in the FLASH to improve their security posture in response to ARTEX and similar agentic SI frameworks: ic3.gov/CSA/2026/261113.pdf
ALT FBI Flash report dated November 13, 2026, detailing threat overview of ARTEX Autonomous Penetration Testing Framework.
My old friend @gadievron is 100% right here - per usual. The tsunami is coming. Now is the time to prepare. As a CISO the time to act is NOW. don’t get any credit for predicting rain - only for building an arc.
Industry, please hear us. Autonomous attacks are here, at scale, breaching many organizations right this moment. Usually, with ARTEX and Cairn, usually through vulnerabilities.
A thread
Heather Adkins (@argvee) and I asked the industry to believe us…
1
4
201
Eric Foster retweeted
The way to combat AI cybersecurity attacks is to use more AI on defense.
Lots of focus on attacker using AI.
What doesn't get as much attention is how much AI is helping with defense.
@coinbase has put it at the forefront of our security work: early access to frontier models with leading developers, agents testing our code, and a lower cost to find vulnerabilities - all of which frees up resources for more work.
AI for defense has a clear advantage - we can give it all the context about the system it's defending.
Like everything else in security, that advantage won't last and can be used against us if we aren't careful - which is why this is such an exciting time to be in cybersecurity!
Eric Foster retweeted
Some of the biggest openings in security have nothing to do with patching
How breaches began, per a 2026 report:
- 31% exploited a vulnerability
- 35% used phishing, stolen logins or impersonation scams
And nearly half now involve a vendor or partner, up 3x since 2024
Full piece from a16z's Asaf Ezra and Steph Zhang on six cybersecurity companies worth building: a16z.news/p/defense-in-depth…
Eric Foster retweeted
For two decades, writing the exploit was the bottleneck in a cyberattack. AI made that step cheap, and patching hasn't gotten faster:
- On average, attacks went from landing 63 days after disclosure in 2018-19 to 7 days before it in 2025
- A working exploit went from a skilled person's month to a few hours of machine time
- The median time to fully fix a known-exploited bug rose to 43 days
a16z's @asaf_ez and @steph_zhang expect a long, expensive worm era, hitting hardest where systems can't update fast: hospitals, schools, small water systems.
"When attacking is nearly free, attackers no longer need to want a target."
Full piece from Asaf Ezra and Steph Zhang: a16z.news/p/the-exploit-was-…
Eric Foster retweeted
Good stuff from the JP Morgan Chase team:
jpmorganchase.com/about/tech…
Redolent of what I wrote about back in March on cyber’s need for speed is as much about the management governance layers vs. only the technical layers.
philvenables.com/post/cybers…
Eric Foster retweeted
In a single year, AI has rewritten the security priority list for CISOs.
Attackers are moving faster. Bain estimates AI compresses a typical cyberattack from about four weeks to about 18 hours, and models keep getting better at finding and exploiting weaknesses.
There's also more to protect, with employees shipping apps they couldn't have built a year ago and agents getting access to source code, documents, and production systems.
In a survey of 100 CISOs at companies with 1,000+ employees, all seven of the most-selected 2027 priorities involve AI.
The biggest jumps since 2026:
- Internal AI agent security (+28 pts)
- AI SOC (+26 pts)
- AI offensive security (+24 pts)
- AI code security (+23 pts)
- AI vulnerability remediation (+22 pts)
Last year's staples:
- Exposure & vulnerability management (-21 pts)
- SIEM / security data (-18 pts)
- Privileged access management (-18 pts)
From a16z's @OwenColby and @steph_zhang on The New Security Agenda: a16z.news/p/the-new-security…
One of the best consumer products I own just got 20% better. Best part - buried lede - is that existing owners can upgrade for only the diff between their original purchase price and the price of the Serene (very fair upgrade offer). Thanks @mehul and the entire @maticrobots team
Quiet just got 20% quieter
Introducing Matic Serene
We’ve reimagined our vacuum stack from the ground up to make Matic quieter than ever
1
8
303
I was fortunate enough to see the BlueChip in person yesterday before it launched. Amazing, important work from @PalmerLuckey and @anduriltech team - incredibly critical dependency many overlook in the mission to reboot the arsenal of democracy.
We already have a congressional mandate to buy American made computers for our sensitive and military needs and unfortunately to date there’s been zero American made options. Time to reindustrialize the homeland, anon.
Accelerate, or die.
BlueChip. Built in America.
America made the first electronic computer to win a war.
Now we depend on our greatest adversaries to build the computers that run our economy and defend our nation.
Today, we bring it home.
1
8
660
Lights out in Austin. First race with @TENEXai on the VF-26.
@HaasF1Team's engineers watch the car. Our analysts watch their network. Every alert investigated in seconds, a named analyst on every decision that matters.
hubs.ly/Q04CbpL-0
#HaasF1 #USGP hubs.ly/Q04CbpR70
We are building a rocket ship - and now we are helping secure one.
Such an amazing milestone and so proud of the entire @TENEXai team. Ridiculously excited to partner with @HaasF1Team!
A stronger line of defence🔒
We’re proud to welcome @TENEXai to the team as our Official Cybersecurity Partner in a new multi-year cybersecurity partnership. 🤝
Together, we’re bringing AI-powered security and human expertise to the fast-moving world of Formula 1.
#HaasF1 #F1
7
190
Eric Foster retweeted
Marc Andreessen says the gap between what AI-native startups and incumbents think AI can do is the widest he’s ever seen:
"The AI-native early stage startups have just a completely different view on what they can do with AI than the typical incumbents in almost any industry, almost without exception."
"AI is just racing ahead in capability even week by week... There are going to be a lot of industries that are really up for grabs in a fundamentally more profound way than maybe has been the case in the past."
"The founders are gonna have a level of superpowers coming from the use of these tools and these technologies that they just simply haven't had in the past... Had I had Astra and Netscape in 1994, like, oh my God. Like, Ben, what would you and I have been able to do at Netscape?"
"Those entrepreneurs are gonna color way outside the lines. We already see this. They're not gonna think of themselves as like, 'I'm just a software company,' or, 'I'm just a hardware company,' or even, 'I'm just an AI company.'"
@pmarca @bhorowitz
Today, a16z Crypto is becoming a16z Frontier.
Since joining a16z in 2013, @cdixon has focused on technologies that look futuristic well before they reach the mainstream. Those themes led to investments in Oculus, Skydio, Zipline, Coinbase, Solana, and Uniswap.
Frontier expands that mandate beyond crypto to back "outer planet" founders who are building new technologies and creating markets that may take years to develop.
Chris joins @pmarca, @bhorowitz, and @eriktorenberg to discuss why crypto has become part of the broader tech landscape, why Frontier fills a hole in the firm's architecture, and why they'd rather be too early than too late.
They also get into why the gap between AI-native startups and incumbents is the widest Marc has ever seen, how AI could pull forward 20 years of frontier technology, why stablecoins may become the way AI agents pay each other, how AI is breaking the internet's business model, and why today's 18-year-olds are about to become superhumans.
0:50 Introducing a16z Frontier
2:49 Smartphone parts to drones and VR
4:47 Backing the "outer planets"
7:11 Good ideas that look like bad ideas
10:43 The JP Morgan photo in Marc's office
15:13 Good tech ideas eventually work
16:53 AI may pull tech 20 years forward
18:18 The widest AI gap Marc's ever seen
22:06 Workers adopt AI before companies do
26:32 Old use cases vs. AI-native ones
30:08 Beyond "product manager brain"
34:45 AI can reverse engineer any hardware
38:23 The case for open source AI
46:36 How AI breaks the internet's model
49:11 VR, homebrew hacking, multiplayer AI
54:00 If you're 18, study "modern alchemy"
YouTube: youtu.be/-9XQMHhpypc
Not an offer or solicitation. None of the information herein should be taken as investment advice; Some of the companies mentioned are portfolio companies of a16z. Please see a16z.com/disclosures for more information. A list of investments made by a16z is available at a16z.com/portfolio.
The Hottest Space in AI Security Is About to Get Scored, and Most Enterprises Are Racing Toward Disaster
Early next year I'm running a structured evaluation of the seven leading AI SOC platforms: TENEX.ai, Prophet Security, Dropzone AI, Conifers.ai, CrowdStrike, Microsoft, and Google. This isn't a vendor bake-off or a sponsored ranking. It's a 20-criterion scorecard covering detection coverage, response time, agentic AI maturity, explainability, model governance, integration, compliance, total cost of ownership, and vendor viability, with critical criteria weighted two to three times and a hard disqualification rule for any platform scoring below six on a critical attribute. The goal is to produce the kind of decision framework enterprises actually need when trusting an autonomous system to make security decisions on their behalf. It'll be interesting to see who comes out on top, and I have my suspicions, but the data will decide.
Make no mistake, this is the hottest space in AI security right now. Agentic SOC platforms promise to triage one hundred percent of alerts in under sixty seconds, something no human SOC has ever achieved, and the funding numbers prove the market believes it. But here's what concerns me. Almost every enterprise I talk to is approaching this technology exactly wrong, and many are heading for a disaster that's entirely preventable. They're buying these platforms before they understand what the technology actually does, and more importantly, when not to use it.
So let me be clear about the boundary. These platforms are transformative when you have high alert volume that overwhelms your analysts, a well-instrumented environment with a mature SIEM or XDR already in place, and clear human authority over containment actions. They're a disaster waiting to happen when your telemetry is garbage, your detection engineering is immature, or you hand autonomous agents the keys to production systems without guardrails, auditability, and explainability requirements written into the contract. An AI SOC amplifies whatever security posture you already have, including a broken one. Feeding an agentic system noisy, incomplete data doesn't reduce risk, it automates your blind spots and gives you false confidence at machine speed. And an opaque agent that can't explain its verdicts will fail your first serious audit, your first regulator conversation, or your first real incident, and it will fail publicly.
There's real risk and threat out there right now, and attackers are moving faster than most defensive procurement cycles. But speed without governance is just a faster path to the wrong answer. Before you sign anything in this category, demand proof of alert coverage, red-team results against their own agents, no-training guarantees on your telemetry, and a contractual explainability standard.
4
172
