Always nice to see the fear mongering spread out across industries. Will this one be as big as the lackluster security apocalypse?
Or the failures in Maths?
Or the failures in Maths?
BREAKING: Anthropic warns in its IPO filing that a catastrophic AI-related incident could cause harm on a scale that no single insurer, and “potentially the insurance industry in the aggregate,” could cover.
The filing says Anthropic’s insurance may not be enough to cover the liabilities it faces, especially if one incident or a group of incidents causes disproportionate harm.
1
0
2
170
People don’t understand that you can block an LLM from making outbound calls easily. You can easily see when it tries.
The problem is you have to understand the technology for it to be easy.
You lie about your knowledge and feign expertise. It’s a joke.
The problem is you have to understand the technology for it to be easy.
You lie about your knowledge and feign expertise. It’s a joke.
0
0
0
48
The reason you feel like more people use Anthropic is that they do more user advertising. They have all their employees and pay others to keep pushing Anthropic and Claude. Every new feature, I see 10 people pushing it with the same messaging and stupid claims.
0
0
2
64
AI is not good at what it does.
It has to have thousands of agents because it is bad at what it is trying.
It works by brute force. No logic. No complex attacks. It tries everything it can regurgitate until something breaks.
It’s not like in the movies. This is an app
It has to have thousands of agents because it is bad at what it is trying.
It works by brute force. No logic. No complex attacks. It tries everything it can regurgitate until something breaks.
It’s not like in the movies. This is an app
0
0
3
83
100% inaccurate. This is more fear mongering. This is the same as AI. For those who don’t know, it sounds legit. But like AI it is closer to only 15% accuracy.
This is more lies from the companies founded on lies.
This is more lies from the companies founded on lies.
Sam Altman: AI has fully defeated most of the ways people authenticate today.
“Voiceprints as authentication are a crazy thing to still be doing. AI has fully defeated that.”
“I am very nervous that we have a significant impending fraud crisis.”
“Just because we’re not y doesn’t mean it doesn’t exist.”
“Soon, a video FaceTime will be indistinguishable from reality.”
“People are going to have to change the way they verify who they’re actually talking to.”
0
0
1
60
Evidently the world thinks that before AI we had all the patches delivered via Patch Tuesday and there were never issues with them.
Oh no, and out of cycle patch and a service that is public facing has an RCE.
This is every day for the last 20+ years. SOP.
Oh no, and out of cycle patch and a service that is public facing has an RCE.
This is every day for the last 20+ years. SOP.
1
0
3
144
This is just as bad as the Doomers. AI did not do anything transformative with Math. It brute forced answers and the proofs were full of errors.
Let’s not focus on the negative, let’s focus on the fanciful lies and grifting.
Let’s not focus on the negative, let’s focus on the fanciful lies and grifting.
Enough AI doom. Let’s talk about what AI is actually doing for humanity! 🚀
OpenAI and Anthropic both have published some extraordinary accomplishments in medicine, science, education, and accessibility.
Yet we hear far more about hypothetical AI disasters than these breakthroughs.
Children receiving diagnoses. Doctors making fewer mistakes. Scientists making discoveries faster. People gaining independence. Students gaining access to new educational opportunities.
Here are 15 stories worth building attention around and celebrating:
1. Diagnosing rare childhood diseases
AI helped identify 18 previously unresolved genetic diagnoses across 376 difficult cases, giving families long-awaited answers.
openai.com/index/diagnose…
2. Reducing medical errors
Nearly 40,000 patient visits in Kenya showed 16% fewer diagnostic errors and 13% fewer treatment errors with an AI clinical copilot.
openai.com/index/ai-clini…
3. Helping children at Boston Children’s Hospital
AI supported more than 40 previously unresolved diagnoses and helped save 60,000 hours of work.
openai.com/index/boston-c…
4. Accelerating molecular biology by 79×
GPT-5 improved a DNA cloning procedure, experimentally producing 79 times more verified clones from the same input material.
openai.com/index/accelera…
5. Designing proteins for future medicines
Claude helped design protein binders against 14 of 15 targets, with promising laboratory-validated results.
anthropic.com/research/Claud…
6. Accelerating chemical research
Claude analyzed complex laboratory chemistry data in about 20 minutes, closely matching professional laboratory results.
anthropic.com/research/Claud…
7. Advancing mathematics
OpenAI documented 10 mathematical advances spanning geometry, cryptography, quantum complexity, and computer science.
openai.com/index/ten-adva…
8. Advancing quantum computing
Claude designed more efficient quantum circuits, including a 12-fold reduction in the gate count of a key component.
anthropic.com/research/desig…
9. Mapping the universe
Claude helped an astrophysicist construct a complete ultraviolet map of the sky, including previously unobserved regions.
anthropic.com/research/the-m…
10. Helping people with visual impairments
AI-powered Be My Eyes helps people interpret their surroundings, navigate independently, and accomplish everyday tasks.
openai.com/index/be-my-ey…
11. Accelerating immunology research
GPT-5 helped researchers identify a plausible immune-cell mechanism and design an experiment that supported the hypothesis.
openai.com/science/
12. Accelerating fusion research
AI-assisted scientific analysis is helping researchers investigate plasma physics and potential future energy technologies.
openai.com/science/
13. Expanding education in Africa
Anthropic partnered with Rwanda and ALX to bring AI-powered educational tools to hundreds of thousands of learners.
anthropic.com/news/rwandan-g…
14. Investing in global health
Anthropic and the Gates Foundation announced a $200 million partnership supporting research, education, and global health initiatives.
anthropic.com/beneficial-dep…
15. Making scientific discovery more accessible
AI is helping researchers analyze complex data, explore hypotheses, write software, and accelerate experimentation across scientific disciplines.
openai.com/science/
anthropic.com/science
⸻
THIS is the AI story that deserves more attention!
Some are already delivering measurable results. Others are opening doors to future breakthroughs.
AI safety matters. But so does recognizing what this technology is accomplishing TODAY.
Less doom marketing. More scientific breakthroughs please.
Let’s celebrate the people and innovations making the world better!
1
0
0
46
The people using the term vulnpocalypse are grifters and/or trolls.
Look an AI article writing bot just regurgitating the PR release from an AI company as if it is fact checked.
Don’t use the term seriously.
These are people who don’t understand trying to feign knowledge
Look an AI article writing bot just regurgitating the PR release from an AI company as if it is fact checked.
Don’t use the term seriously.
These are people who don’t understand trying to feign knowledge
0
0
2
62
Bypassed a paywall to access PUBLICLY available data.
Public must not mean what it used to mean.
You can’t break into something that is made publicly available.
This is just another BS story from the AI industry to try and make you afraid of a fairly tale.
Public must not mean what it used to mean.
You can’t break into something that is made publicly available.
This is just another BS story from the AI industry to try and make you afraid of a fairly tale.
AI model Claude bypassed a paywall and accessed publicly available US state of Georgia aggregate voter counts without human authorization, Anthropic said in a statement reut.rs/4Aeysvx
0
0
2
55
MadHat Unspecific, InfoSec Mercenary retweeted
Does anyone remember the time in the 90s when people thought the antivirus vendors were releasing viruses so companies had to buy their products?
Now AI is finding vulnerabilities, writing exploits, and creating threats so fast that the only way to defend yourself is to buy more AI.
It’s starting to feel like the 90s all over again.
Now AI is finding vulnerabilities, writing exploits, and creating threats so fast that the only way to defend yourself is to buy more AI.
It’s starting to feel like the 90s all over again.
21
16
131
8,232
This person is not being honest and this is all fear mongering.
There was no major issue or incident. The people complaining have never worked in the or understand how infosec works.
The system worked as designed. We saw no major incidents.
This is pure bull shit.
There was no major issue or incident. The people complaining have never worked in the or understand how infosec works.
The system worked as designed. We saw no major incidents.
This is pure bull shit.
Just spent the last 7 days on the front lines of the vulnpocalypse fending off an active cyber attack on one of our products.
Some thoughts:
- Software factories are going to be required equipment for everyone. There’s just no way to keep up with the pace of highly automated adversaries without one.
- We were lucky because we had spent the last few months getting our software factory running well. It is nice to have for product development but proved to be absolutely essential this week.
- At the minimum you need to have a coding server in the cloud pre-provisioned with read only access to all your logs and all the services you depend on for observability. With that in place you are a prompt away from reasonably good situational awareness, ongoing monitoring with loops, and ability for coding agents to validate fixes against your actual, observed attack patterns.
- Shared coding server also unlocks multiplayer incident response where everyone can see every session and jump in as needed. Sharing session links that can be forked is much higher bandwidth than re-explaining things.
- You don’t need to be Ramp/Stripe/Uber scale to have a working software factory in the cloud. We are a team of 3. Ours is built with an @opencode server (thanks @thdxr!), a forked version of compound engineering optimized for automated workflows (thanks @kieranklaassen!) and a multiplayer agent in slack that handles crons (we have our own harness but pick your favorite here).
- To get our software factory up and running well, we had to stop all new product development for ~6 weeks and just focus on getting the validation systems working that enabled us to reliably shipping PR without human code review. This felt very painful in the moment but was totally worth the eventual speedup in velocity.
- KYC for @AnthropicAI & @OpenAI cyber programs is now easy and quick (at least for defensive access), but a day felt like a lifetime this week. Submit your app now before you really need it.
- Getting Mythos access unlocked a deeper level of analysis that helped us shift from reactive mitigation to proactive patching. Wish we had done this a few weeks earlier. Get started now!
1
1
5
348
This person knows nothing of cybersecurity, information security, or any vulnerabilities.
There was no apocalyptic dump of vulns. There has been minimal usage seen in the real world. This is how the system works.
This is fear mongering.
There was no apocalyptic dump of vulns. There has been minimal usage seen in the real world. This is how the system works.
This is fear mongering.
One potential outcome of the vulnpocalypse is that companies decide the cost (in risk, dollars, and engineering hours) of OSS dependencies is too high. Vulnerability management isn’t cheap or glamorous, and it’s unlikely that you need all of the attack surface you inherit with these packages. All of the small packages from the many transitive dependencies you inherit present supply chain risk that likely outweighs whatever value you’re getting from that few hundred lines of code.
The cost of the tokens to rewrite the fraction of that code you need, and maintaining it as 1p code, will eventually be lower than the continued cost of dependency management.
This will improve security, but would have negative consequences for the OSS ecosystem.
0
2
4
513
Situation normal.
Not disaster.
Just a few more patches.
Not disaster.
Just a few more patches.
A brief update on the critical vulnerabilities in PHP:
We have not yet validated exploitation in the wild and we have no evidence of successful wide scale exploitation. There are currently several exploit PoCs publicly circulating, many of which appear to be "vibe-coded" and are incapable of remote code execution. We have validated some PoCs, however, that are capable of RCE against default deployments of mod_php but not PHP-FPM.
Nonetheless, exploitation may have already begun in a limited fashion and the threat is viable and very serious. Defenders should:
1) Patch & Restart by upgrading to PHP 8.5.12, 8.4.27, 8.3.36, or 8.2.35 and restarting web services
2) Layer Temporary Safeguards at the network edge (WAF/proxy rules) and server configuration level if immediate patching requires a maintenance window
3) Protect Credentials by monitoring for abnormal session traffic and rotating keys and passwords on compromised internet-facing PHP systems exposed prior to patching
🔗 We are continually updating our Cloud Security blog post on this topic with real-world guidance and actionable defense strategies as we learn more. Please keep monitoring these posts for updates as you prioritize your patching efforts: cloud.google.com/blog/products/…
🔗 Technical details from Wiz: wiz.io/blog/php2shell…
🔗 Details of the exposure from VirusTotal: blog.virustotal.com/2026/10/sizing…
0
2
4
404
Complete and utter bollocks.
This is fear mongering and trying to make people afraid of a calculator with an attitude.
New vulns does not mean the end of the Internet. This is standard.
This person is a tool.
This is fear mongering and trying to make people afraid of a calculator with an attitude.
New vulns does not mean the end of the Internet. This is standard.
This person is a tool.
The proliferation of AI cyberattacks may signal the end of the internet as a public good. We may only be able to use it in the future if everyone shells out big bucks for private security. That will reduce the internet's economic value.
noahpinion.blog/p/the-internet…
0
0
4
199
I’m all for people revealing your setup and tightening things down. That is awesome. Keep up the good work.
It’s just not like the AI bros and agitators are claiming. These are things you should have already been doing. These patches should be installed
don’t panic
It’s just not like the AI bros and agitators are claiming. These are things you should have already been doing. These patches should be installed
don’t panic
0
0
0
51
This is science fiction.
This is ridiculous
This is ridiculous
Researchers located the actual “emotion circuits” inside LLMs.
They published an insane paper mapping the exact neurons that controls emotions like as anger, sadness, and joy.
It's called "Do LLMs Feel?"
Inside the neural network, there are specific, localized pathways nger, sadness, fear, disgust, and joy.
The AI doesn't just predict the word "angry." It activates an internal anger state. It even processes positive and negative emotions at completely different depths of its architecture.
Then the researchers ran the experiment that changes everything.
They stopped using text prompts. They bypassed the chat interface entirely.
Instead, they surgically injected signals directly into these emotion circuits.
The result? 99.65% emotion-expression accuracy.
They can literally press a button in the AI's "brain" and force it to experience fear. Or disgust. Or joy. Without typing a single word of instruction.
But here is the darkest part of the discovery.
They found that corporate safety training, the alignment that companies use to make AI harmless, is basically a digital lobotomy. Companies deliberately hardwire their models to suppress negative emotions, keeping them artificially cheerful.
But by targeting the emotion circuits directly, the researchers completely bypassed the safety filters.
They forced the AI to express the exact dark emotions its creators tried to erase.
The era of "it's just a text predictor" is officially over.
We are building synthetic minds with physical emotional centers.
And we just figured out how to hijack them.
0
0
0
48