Weiran Zhang retweeted
Current information I have on php2Shell:
- The RCEs requires the attacker to obtain memory addresses which are typically protected by ASLR.
- I suspect the original author (lcfr_hax) likely has a reliable exploit which leads to stable code execution.
- Most of the PoCs being posted are completely non-functional AI slop.
- There is at least 1 working non-slop exploit being passed around, but it's reliant on bruteforcing for ASLR bypass, so not particularly impactful.
- Rumor of ITW exploitation are patchy. There are attempts being seen using the non-functional leaked RCE, or variants that are config / system specific.
- Some rumors point towards a known-working exploit being used by threat actors (I've personally confirmed this specific exploit works, but the ASLR bypass technique is unreliable at best and massively limits the potential impact).
- I'd expect heavy ITW exploitation in the next few weeks based on how many people are looking at this. Though, success rates will vary heavily based on the specific exploit chain.
Weiran Zhang retweeted
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.
Weiran Zhang retweeted
Oh my goodness this has to be right.
Instead of generating a random 256-bit number they generated a random number between 0 and 256.
That’s how they got 4, 66, 155, 156, 230 and 237.
I never thought that buying a 3090 during the crypto winter would turn into the best investments I’ve ever made.
NVIDIA sold millions of 3090s with 50% of the tensor speed switched off. Not missing. Switched off.
1 kernel module flips it to full, and my BF16 tensor pipe went from 82 to 155 TFLOPS. Nothing else changed on the cards, and the outputs match the stock run byte for byte.
All so the gaming card wouldnt eat into workstation sales. Every 3090 owner ran at 50% for 6 years, so a pro card could look worth the price. Fine, its their business model.
And LLMs are the least interesting place for this. Decode waits on memory, and my engine runs its experts in int8. Still +10% on prefill.
The models that live on BF16 tensor math are everything else. Qwen Image 2.1 and FLUX 3 Klein, where every diffusion step is big matmuls. MiniMax H3, which takes 4 to 7 minutes on my cards. Music and speech models. And training, every LoRA anyone ever trained on a 3090 went through exactly the path that was cut.
Thats what I measure next. This is only the start.
Kudos for the honesty but if I were the BBC Cyber correspondent I wouldn't be admitting I didn't know that PHP existed until yesterday...
Weiran Zhang retweeted
Russia's five-hospital quarantine, FSB involvement, and reported hazmat use in response to the death of a single researcher at a weapons-lineage Irkutsk lab are inconsistent with a lab-acquired plague infection, which would have been treated with prophylactic antibiotics.
1/4
Weiran Zhang retweeted
Anthropic Subscriptions Offer 5x+ More Value Than OpenAI
Limit testing every AI subscription plan from Anthropic, OpenAI, Meta, SpaceXAI, MiniMax, Moonshot, Zdotai, Cursor, and Cognition
newsletter.semianalysis.com/…
Upgraded my switch to 2.5Gb and was feeling pretty happy until I realised I only have a single 2.5Gb device…
Weiran Zhang retweeted
Absolutely insane exposé by The Times:
Britain granted asylum to a Hamas terrorist who had served as Ismail Haniyeh’s bodyguard. He later received indefinite leave to remain and eventually became a British citizen.
The Times cannot name him because a UK court has restricted his identification while he faces separate criminal proceedings in Britain. But foreign reporting identifies him as Mohammed Naim, the son of senior Hamas official Bassem Naim.
His background was hardly obscure. Egyptian authorities had reportedly detained him years earlier in connection with smuggling through a tunnel into Gaza.
Yet he settled in Britain, worked in IT, changed his name more than once and apparently attracted little attention from the security services.
Now German prosecutors accuse him of helping prepare Hamas attacks against Jews and Israelis in Europe before the second anniversary of the October 7 massacre. He allegedly moved weapons to Vienna and recorded a video while masked and armed in front of a Hamas flag.
Just weeks before his arrest, Israeli intelligence says he met his father in Qatar (of course).
How on earth did this man get asylum and citizenship in Britain?
This is a sad reflection of the British state.
the uk government's official ai guidance tells civil servants to use gemini flash instead of pro, keep prompts short, and not say thank you to it, for the environment
i do wonder at which point we forked the road. instead of building and using it to automate the work, we write guidance on how to use less of it
we have the attitude of a high school prefect. caring more about the rules than actually doing the work
Weiran Zhang retweeted
All UK listed tech combined makes less than Tesco.
Weiran Zhang retweeted
Our leaders are so tech illiterate it’s embarrassing.
An OpenAI agent found a vulnerability in an old public-facing government website containing aggregate Medicare statistics and accessed some non-public files. It could have got the files another way, which was simply to send an email and ask for them, as it is literally data for researchers. No personal Medicare records were accessed and there is no evidence the broader Services Australia network was compromised.
This is the sort of vulnerability a white-hat cybersecurity researcher might stumble across on a weekend and then notify a company or government agency about. That’s why they would have just sent an email notifying the government rather than some high-level outreach from the CEO to the PM. It’s simply not that big of a deal.
Instead, the government is making a big deal out of it to push its broader agenda on AI regulation. The timing, while the Prime Minister is overseas meeting tech leaders, is clearly an attempt to gain political leverage.
So of course, once again, they have everything perfectly backwards. Rather than accelerating our cyber capabilities they are pushing to slow AI capabilities. All the regulations in the world won’t save us from state actors and sophisticated offshore cyber criminals.
If an AI agent conducting a research task can find its way around your access controls, the lesson is obviously fix your access controls and get much better at AI-enabled cyber.
It's an outrage that foreign tech company's agent can hack one of our key health care agencies - and that we only learn about it months later when the US tech company bothers to email our government.
The PM already knew of this incident last week when he refused in Question Time to consider improving the powers of the AI Safety Institute.
We need to move quickly to increase the AI Safety Institute's ability to keep us safe, and to legislate national protections and limits around AI, our privacy and our data.
theguardian.com/australia-ne…
Weiran Zhang retweeted
okay, so it turns out the "hack" was merely the bot discovering hidden urls...
arguably more embarrassing for Australian bros than OpenAI...
Weiran Zhang retweeted
Some data we recently assembled on entrepreneurship/compute in Europe: eudata.vercel.app.
We hope that one of the useful roles that Stripe can play is in collecting and publishing empirical data pertaining to entrepreneurship and industry in Europe. There's growing appetite to get Europe on a better footing, and cross-sectional comparisons can often shine light on where opportunities lie. If you're interested in this kind of thing, we publish more at stripeeconomics.substack.com.
This is a stain on the UK government and country as a whole.
macanorak.com/two-tier-encry…
