ZeroHour

Search: “Direction”

21,147 stories

How Fragile Is Safety Alignment at Frontier Scale? A Single-Direction Attack on a 320B MoE

Researchers show directional ablation breaks refusal in GLM-5.3-Flash, a 320B-parameter MoE, cutting refusal by 41–89 points across seven benchmarks.

The study extends directional ablation, a white-box attack that removes an aligned LLM's refusal behavior, from dense models up to ~70B parameters to GLM-5.3-Flash, a 320B-parameter mixture-of-experts model with 288 routed experts, four-wide hyper-connection residual, and block-FP8 quantization. Editing attention, dense, and routed-expert writers jointly removes 0.776 of refusal, with 74% of the effect existing only under the joint intervention; the conventional module-name-based recipe reaches only 0.066 and fails silently on MoE architectures. The attack yields 41–89 percentage-point reductions in refusal across seven harmful benchmarks with no detected capability change, and a category-concentrated refusal residue survives all edits at ranks 1 to 12.

arXiv cs.CR · 7d agoAI safety & security

Payara 7.2026.1.RC1 Remote Code Execution via Server-Side Includes #exec Directive in Payara Server

Payara Server 7.2026.1.RC1 executes arbitrary OS commands when user-controlled Server-Side Includes #exec directives are passed to Runtime.exec without validation.

Payara Server contains a vulnerability in its Server-Side Includes (SSI) implementation that permits arbitrary operating system command execution via the #exec directive. User-controlled SSI directives are passed directly to Runtime.exec() without validation, sanitization, or restriction. An attacker who can cause the server to process an SSI file such as .shtml can execute arbitrary OS commands. The disclosed affected version is 7.2026.1.RC1.

Full Disclosure · 12d agoVulnerability 2 sources

You don’t have to join the hack-back program to inherit its risk

A new US presidential memorandum creates a vetted private hack-back program, leaving participating vendors and their customers with untested legal liability and collateral risks.

The August 12 National Security Presidential Memorandum directs the National Coordination Center, run jointly by DOJ and DHS, to approve covert surveillance and disruptive Cyber Effects Operations by vetted private companies, with a forfeitable bond of at least $1 million required as a contract condition. The analysis argues the criminal shield rests on an untested reading of the CFAA exemption at 18 U.S.C. 1030(f), with no civil safe harbor, no state-law preemption and no foreign-law protection. Non-participating organizations can still inherit risk through shared infrastructure collateral damage, lack of customer disclosure, Lloyd's bulletin Y5381 state-backed attack exclusions, and threat-intelligence pipelines feeding offensive proposals.

CSO Online · 10h agoPolicy & legal

The complex corporate web behind a $3.2 billion AI data center

Ars Technica probes diffuse accountability behind TeraWulf's $3.2B Lake Mariner AI data center after a June fire exposed safety and job gaps.

A June fire at the Lake Mariner data center in Somerset, New York exposed missing alarms, a nonfunctioning suppression system, and dry hydrants, highlighting how responsibility is split across TeraWulf (owner-operator), Fluidstack (operator), Google (lease guarantees and equity warrants), and Anthropic (compute customer). The article details local concerns over the gap between promised 165 permanent jobs and a projected 35-40, socialized grid costs, and Governor Hochul's moratorium on hyperscaler development. Anthropic's February 2026 pledge to cover electricity price increases applies to the site but leaves other commitments unverified.

Ars Technica · AI · 9d agoAI industry