Building Security + Enterprise Products @CoreWeave. Built Security Products @HashiCorp, Crypto and Defense Products @NetApp. š¤: Kendo, trance music, Warhammer
NYC Metro Area
Joined August 2008
- Tweets17.6K
- Following664
- Followers2.1K
- Likes22.4K
Pinned Tweet
The traffic sucks. Everything's expensive. But living in a place where convos abt physics, CS, and math are the daily norm rules.
Hot take:
Explosive days in security like today are because AI is showing us that the night is darker and more full of terrors than we imagined.
If we donāt build widespread agentic defenses and remediation, itās going to get very bad. Very fast.
Andrew "Andy" Manoske retweeted
Regarding potential attacks on energy, food, and water infrastructure, the threat is quite real. Someone is currently probing infra and feeling it out, it is an active threat. Even a little preparation can be helpful, just in case. This comes from the joint cybersecurity advisory released by the NSA and partner agencies on August 19:
'The threat actors are conducting reconnaissance and capability development against U.S.-based Siemens PLC installations using Al-generated exploitation scripts disguised as legitimate monitoring tools. The actors leverage Internet scanning services to find Internet-exposed PLCs running outdated software or that are otherwise poorly protected. The U.S. critical infrastructure sectors most targeted by this threat activity include Critical Manufacturing, Energy, Water and Wastewater, Chemical, Food and Agriculture, and Commercial Facilities. This is not a theoretical risk - it is an active threat. Depending on the specific circumstances, exploitation of poorly protected PLCS could lead to disruption of critical industrial processes, safety incidents, downtime or equipment damage, compromise of sensitive data, compliance violations, and cascading impacts across interconnected systems.'
'Using Al to generate exploitation scripts represents an evolution in threat actor capabilities, dramatically reducing the technical expertise and time required to develop working ICS exploitation scripts and malicious tools. In addition, Al enables adversaries to rapidly leverage additional attack vectors and adapt to defensive measures. Threat actors can easily collect public information about vulnerabilities and weaknesses, find exposed and exploitable PLCs, and use Al-generated scripts to act on that information. If PLCs are exposed to the Internet, they are at high risk for exploitation.
Threat actors are leveraging open source industrial automation libraries - specifically snap7.d11/python-snap7 - combined with Al-assisted scripting to create custom tools that mimic legitimate OT monitoring solutions. These tools provide read/write access to Siemens S7 Series PLC memory, configuration data, and ladder logic programs via the S7comm protocol.'
It seems like we're in a pretty hot moment cybersecurity wise š¬ If you're reading this website, you probably know more than most of the people in your life about what's happening! they're not talking about any of this stuff on the news, it's too arcane and weird and sudden. It is on us.
Good time to call or text your family and friends, have some cash on hand, have a meetup point in your neighborhood for if the Internet goes down entirely, etc.
The full prepper mode for shit hitting the fan is probably something like what Andreas posted here but you can get 80% of the way with 20% of the effort. Think about what that would mean for you in your situation, and the people you care about, and talk to them about it sooner than later.
And as always, the most important part of prepping is having a strong support network and caring for each other. "Going off grid" isn't the move, it just makes you isolated and you have to do everything yourself. We are social animals. Community is our superpower š«”š«”š«”
One word: agents.
Some stats on #php2shell. In the 10 hours before it was publicly announced, across the @Cloudflare network we saw thousands of attempts to exploit it. That jumped to tens of thousands in the 45 minutes post announcement, hundreds of thousands within 2 hours, and millions now.
Andrew "Andy" Manoske retweeted
OpenAI models helped researchersāoperating as part of our Patch the Planet initiativeādiscover critical security flaws in PHP, an open-source programming language used to run an estimated 70% of websites.
Over the past two months, we led a coalition of maintainers, researchers, and security providers to prepare fixes and protections ahead of disclosure. This duration was necessary because the fix was quite intricate and required a major change to the PHP engine. Every site using PHP can now safely upgrade their version to stay secure, in addition to applying mitigations from various security vendors.
Due to the severity of these flaws and PHPās widespread use, we consider this bug to be one of the most impactful vulnerabilities in recent memory. It also marks a new phase, where AI can find security bugs with internet-scale impact.
This is an example of the Defenderās Window in action: to our knowledge, frontier AI was able to find this vulnerability ahead of attackers, giving defenders time to prepare.
openai.com/index/coordinatedā¦
Readers added context they thought people might want to know
āŖOpenAI may claim a role in the coordinated response and patching process. But the initial discovery and exploit development, however, occurred several months earlier by an independent researcher.ā¬
āŖx.com/lcfr_hax/statuā¦ā¬
Andrew "Andy" Manoske retweeted
Crazy 72 hours in security. Three major incidents dropped back-to-back: eMudhra, PingCAP (TiDB Cloud), and PHP.
Here is the short version of what happened and who got hit.
1. eMudhra (Trusted CA compromise)
eMudhra generated 8 intermediate CA certificates in April 2025 (cross-signed by Sectigo). A bug in their HSM key-wrapping test script created a second set of P-384 keys where the private keys were single-byte numbers under 256. Operators picked the wrong key labels and signed them into real CA certs.
One of the intermediate CA private keys was literally the number 4.
The other seven: 56, 66, 115, 155, 156, 230, 237.
Three weeks later (May 2025), their CRL signatures started failing because the CRL signer was using the real keys instead of the weak ones. Instead of asking why the keys didn't match, someone switched the CRL signer to the weak keys and closed the ticket.
ā¢ Browsers are safe: Chrome, Firefox, Apple, and Microsoft already blocked them, and zero rogue certs hit CT logs.
ā¢ Servers are not: curl, Python (certifi), Go binaries, and Docker base images don't check CRLs. Update ca-certificates, certifi, and Go (1.27.3 / 1.26.10).
ā¢ 10-second check on any server:
curl dv-g3b-emudhra.we-have-a-proā¦
(TLS error = safe. 200 OK = still vulnerable.)
2. PingCAP / TiDB Cloud (10.6 TB breach)
Attackers found GCP and Azure credentials committed inside PingCAP's own repos. They set up a replication feed of TiDB Cloud's production control plane and exported the entire database every 6 hours for 22 straight days. 88 full exports, zero alerts. 200 GB is already out on torrents.
Beyond the 413 private repos and 20.6M customer SQL queries logged from their web SQL Editor, PingCAP engineers were pasting customer secrets into Jira tickets.
Who is affected in the leaked records:
ā¢ Live credentials exposed in support tickets:
- Pinterest: AWS keys for their pingcap-logs S3 bucket
- CardX / Silverlake (NPCI-linked env): AWS admin key in a backup log with full access to 258 external S3 buckets
- DeBank: 38 presigned S3 URLs
- An unnamed AI/dev platform: a 314 MB slow-query log attached to a ticket containing 8,027 live customer secrets (an AWS root key, a GitHub admin PAT over 1,701 repos, 7,485 AWS STS tokens, and 192 OpenAI/Anthropic/Gemini keys)
ā¢ Tenants with infra metadata, VPC CIDRs, app-server IPs, or contract pricing exposed:
- Tech: Atlassian, Databricks, Airbnb, Baidu, LINE, Nylas, Persona, PatSnap
- E-comm & Travel: Booking.com, Shopee, Flipkart, Mercari, ZOZO, AEON, Delhivery, Klook
- Gaming: Capcom, Square Enix, Konami, Bandai Namco, Cygames, Colopl, DMM
- Finance: PayPay, SoftBank, SBI, SB Payment Service, Fujitsu, TOPPAN, Bank of China, China Life
- Crypto: Bitget (said only public trade logs were on it and they already left), Pionex, DigiPlus, NFTscan, Chainbase
ā¢ Developer profiles leaked (from a purchased reo.dev table in PingCAP's BigQuery):
- 24,900 personal emails + GitHub/LinkedIn profiles of engineers at Tencent (3,327), Huawei (2,726), ByteDance (2,475), AIG (2,140), Mastercard, Flipkart, and bKash.
If you use TiDB Cloud: rotate your console API keys, DB passwords, and anything you ever pasted into a support ticket.
3. PHP "PHP2Shell" (CVE-2026-109618 & CVE-2026-109619)
Two CVSS 9.2 pre-auth RCEs leaked yesterday ahead of today's 14:00 UTC patch.
Public writeups trace it to a heap Use-After-Free in PHP's native C multipart/form-data session upload-progress handler (ext/session/session.c).
Because C-level multipart parsing runs before line 1 of index.php:
ā¢ App auth, Laravel/WordPress middleware, and file_uploads = Off do not stop it.
ā¢ Only PHP 8.2ā8.5 get official patches today (59% of PHP servers on Shodan are on EOL <= 8.1).
ā¢ Until you patch php-fpm: lock php_admin_flag[session.upload_progress.enabled] = off and block unexpected multipart requests at Nginx/Apache/Cloudflare.
Andrew "Andy" Manoske retweeted
Bringing production-scale AI inference to life takes a strong open ecosystem. As a founding member of the project to now releasing llm-d 1.0, we're thrilled to support this community and fuel the future of open innovation šŖ
Scale distributed inference and control AI serving costs with confidence. llm-d 1.0 brings multi-tenant quality of service, zero-downtime rolling updates, and prefix-cache-aware routing to your production workloads. Explore the release: red.ht/4ysmUU5 #PyTorchCon
Andrew "Andy" Manoske retweeted
Everything comes down to human bottlenecks right now, or will come down to them in the next 12 months. And the defender bottlenecks are awful. Thereās going to be a lot of pain before that changes.
The Agentic Adversary is out and absolutely stunting on traditional infrastructure.
We need better agentic defenses. Yesterday.
HOLY WTF. Reuters is reporting that a Chinese-speaking hacker used AI agents to compromise websites belonging to more than 100 companies, including Deere, SharkNinja and Cummins, after attempting intrusions against over 1,200 sites.
The estimated cost: roughly $5,000 in AI tokens, or $50 per successful breach. Successful intrusions took around 23 hours on average.
The attackers used the open-source Cairn agent framework, apparently combining DeepSeek models with a Claude Code harness. One campaign gave its agents instructions in fewer than 100 Chinese characters.
NorForge researcher Snorre Fagerland: 'It's a new world now.'
Andrew "Andy" Manoske retweeted
So this is cool. If you ever need an ephemeral dev environment and need to scale up resources try out running k3s in our @CoreWeave sandbox platform. Using private networking via @Tailscale we can spin up a k3s cluster with many worker nodes. Perfect for local e2e testing or dev environments.
Anthropic discovered the original NIST seeds for ECDSA!
ā
Disproves 30yr+ rumor that NSA weakened ECDSA using non-random seeds
ā¼ļø Proves that Anthropic (and likely OAI) unreleased models capable of nation-state level cryptanalysis and codebreaking
anthropic.com/research/ecdsaā¦
So does this mean some ācurious engineerā just won $12k?
I bet next year some MTS 2 at Anthropic or OAI is going to buy out the floor tables of XS for Knock2 based solely on their agents tearing through a week of HackerOne and crypto bounties.
Everyone wants to be gangsta until it's time to [create highly available physical infrastructure capable of continuously hosting hyperscaler-class HPC workloads despite half of it being on fire or exploding].
It's hard to see the results from GTIG and not think that the using the following need serious foundational changes in the modern, AI era:
- NOCs/SOCs, way more than "Agentic SOC"
- Threat Intelligence sharing and use
- SIEMs
- IDS/IPS platforms, especially rule management
If you're an enterprise without agentic defenses going into the late '20's, you're fighting full auto machine gun-armed adversaries with a single-shot musket.
Most successful cyberattacks in '27 are agents simply outrunning human-instrumented defenses.
cloud.google.com/blog/topicsā¦
If you're an enterprise without agentic defenses going into the late '20's, you're fighting full auto machine gun-armed adversaries with a single-shot musket.
Most successful cyberattacks in '27 are agents simply outrunning human-instrumented defenses.
cloud.google.com/blog/topicsā¦
Exactly.
Really hard problems get easier to solve. They don't go away.
Given
ā
Nvidia RTX Spark Laptops
ā
Apple Silicon CPU/GPU arch
ā
AMD APUs
ā
Intel Panther Lake going full chiplet
ā
Every other performance ARM chipset being a chiplet
It seems like discrete GPUs on the way out for mobile computing, even in gaming and performance workloads