Smart search ranks by meaning as well as keywords (one row per story, last 45 days).
Multiverse Computing's Hugging Face post argues language models should refuse only the relevant subset of a topic instead of over-refusing whole subjects.
A Hugging Face blog post by Multiverse Computing examines refusal granularity in language models, arguing models should refuse the relevant subset of a topic rather than the entire topic. No full article text was available for additional technical detail.
Stub page for Palo Alto Cortex XDR with no article text, offering no news content for classification.
The item contains only the title 'cortex xdr' and no article body. It references Palo Alto Networks' Cortex XDR product but provides no event, release, or threat information. Classified as general vendor/industry content with minimal relevance.
Palo Alto Unit 42 Cortex Xpanse product-category page published with no article text available for classification.
The URL resolves to the Cortex Xpanse product category listing on the Unit 42 blog rather than a research article. No article body was provided, so no vulnerabilities, incidents, or research findings can be extracted.
Vendor product-category page for Palo Alto Unit 42 IoT security content, published with no article text available.
The URL points to Palo Alto Unit 42's 'IoT Security' product category listing rather than a research article. No article body was available, so no incident, vulnerability, or research findings can be extracted from this item.
Opinion essay titled Merchants of Insecurity, apparently critiquing security industry practices; no article body was included in the feed.
The feed provides only the title and a comments link for this Lobsters-shared post from blog.happyfellow.dev; no article text was available. Based on the title, it appears to be an opinion piece about incentives in the security industry, possibly arguing vendors profit from perpetuating insecurity.
Palo Alto Networks Unit 42 published a threat bulletin, but no article body was available for detailed analysis.
The feed item contained only the title 'threat bulletin' with no article text. No specific incidents, actors, or vulnerabilities could be extracted from the available content.
Training-free reasoning-plus-retrieval framework lifts multilingual multimodal entity linking accuracy by 6.9% overall and up to 23.3% on rare entities.
The paper broadens rarity measurement in multimodal entity linking using knowledge-graph structural metrics beyond popularity-based pageview metrics, identifying many rare entities that popularity metrics miss. Across these rare-entity slices, state-of-the-art accuracy drops by 15.4-39.9%, exposing distinct failure modes under different rarity definitions. The proposed training-free framework iteratively searches and reasons over Wikipedia with a reasoning-capable vision-language model; experiments show reasoning and retrieval are complementary. On the MERLIN benchmark covering Hindi, Indonesian, Japanese, Tamil, and Vietnamese, the system improves overall accuracy by 6.9% and up to 23.3% on rare entities, with MERLIN-Rare test slices released.
CiteShade attack makes RAG models cite trusted sources for attacker-chosen wrong answers, raising wrong-answer rate from 0.01 to 0.68.
CiteShade is presented as the first citation laundering attack against multi-source retrieval-augmented generation: an attacker controlling a single source induces a wrong answer falsely attributed to a trusted source, even while correct evidence remains in context. The attack is formalized via three necessary conditions (retrieval, generation, citation) constructible without any instructions, raising wrong-answer rate from 0.01 to 0.68 on multi-hop QA, with source deletion confirming the malicious source as causal driver. Vulnerability tracks a model's citation propensity rather than scale, reaching CLR 0.84 with explicit instruction and 0.64 without on the most citation-prone model. Perplexity filtering and citation-support checking prove insufficient; the authors propose a counterfactual defense verifying which source actually drove the answer.
SCPTM graph-augmented variational topic model shows syntax aids topic diversity and descriptor quality but gains stem mainly from the variational encoder.
The Structural Contextual Probabilistic Topic Model represents corpora as heterogeneous document-word graphs with lexical and syntactic edges processed by a Graph Attention Network inside a VAE for mixed-membership topic distributions. Across four corpora, neural gains in document-topic alignment are attributable to the variational encoder rather than syntax, while graph-augmented variants improve topic diversity everywhere. Dependency paths add value on argumentative deliberative texts but are redundant in technical and institutional registers.
Zero Day Initiative disclosed an uncontrolled search path flaw in pdfforge PDF Architect's update service, allowing local privilege escalation (CVSS 7.8).
ZDI published advisory ZDI-26-615 describing a local privilege escalation vulnerability in the activation-service Update Service of pdfforge PDF Architect. An uncontrolled search path element lets local attackers escalate privileges, but they must first be able to execute low-privileged code on the target. ZDI rated the issue 7.8 and tagged it as a 0-day disclosure.
Security Affairs malware newsletter Round 111 roundup; the provided page text contains only cookie-consent boilerplate with no story details.
Security Affairs published Round 111 of its malware newsletter, a periodic roundup of recent malware coverage. The provided page text contains only cookie-consent language, so no specific stories, actors, or incidents can be extracted.
Interpretability study traces how Qwen, Llama, and Gemma route query information and internal knowledge across layers when answering questions.
Researchers used layerwise interventions on hidden states to separate query-routing signals from target knowledge in language models, testing Qwen, Llama, and Gemma on country-continent questions with varied answer types. A pair-conditioned request direction strengthens before interventions alter downstream knowledge, opening a causal window while answer-supporting content is still forming. Trajectories differ by model: Gemma shows a partially overlapping mid-layer routing profile, while Llama has no sustained routing-effect window under the same gates.
Schneier on Security posts about Cliff Stoll's DEF CON talk; the visible text provides no substantive details about its content.
Schneier on Security published a brief post referencing Cliff Stoll's talk at DEF CON. The available text consists almost entirely of standard blog navigation, author biography, and archive listings, with no concrete details about the talk itself. As a general security-community item, it carries no direct risk impact for defenders.
404 Media podcast covers Amazon destroying scanned books for AI training, recurring AI names in academic papers, and ICE voter-data spending.
404 Media's podcast follows up on its investigation of an Amazon warehouse where books are scanned and destroyed for AI training data, including an interview with a warehouse worker. The hosts also discuss how the same few names repeatedly surface in LLM outputs and AI-generated academic papers. The episode additionally covers ICE's plans to spend millions on voter fraud data and Boston Dynamics robot dogs.
Dark Reading scheduled a virtual event on building enterprise AI security strategies; no article details were available.
Dark Reading is promoting a virtual event titled 'Building a Secure AI Strategy for the Enterprise.' No article text was available, so details such as date, speakers, and agenda are unknown.
A security researcher revises his stance on AI guardrails, arguing defenders need more help as attackers ignore safety rules.
Dark Reading publishes an opinion piece in which a security researcher changes his mind about AI guardrails. He acknowledges guardrails are critical, citing recent high-profile incidents, but argues defenders need more support because attackers do not play by the same rules. The piece contributes to the ongoing debate on defensive AI controls rather than reporting a new incident.
Opinion piece argues the recent AI warning letter identifies a risk window but omits which actors pose risks and who can mitigate.
This Dark Reading commentary critiques a recent AI warning letter for correctly identifying an approaching risk window while failing to name who is coming through it or who will close it. The piece is brief opinion commentary on AI risk discourse rather than a technical report.
Science article examines expert disagreement over whether a report on AI-enabled bioweapons risks is a chilling warning or an overreaction.
A Science.org article, shared on Hacker News with 20 points and 2 comments, covers expert divisions over an AI bioweapons report and whether its warnings are justified or exaggerated. The discussion reflects ongoing debate in the AI safety and biosecurity community about assessing AI's role in biological threat enhancement. Minimal detail is available from the item itself.
Troy Hunt's weekly update argues AI's hacking role is overstated, citing exposed keys and human error behind recent breach headlines.
Troy Hunt's Weekly Update 521 argues that media coverage exaggerates AI as a hacking tool. He cites data showing AI has had roughly 0% impact on cyber insurance payouts, a MAG breach teardown attributing the incident to exposed keys rather than sophistication, and a Dutch suspect being tracked via his voice. He also notes an Australian headline claiming ChatGPT hacked a court system was actually a man who used ChatGPT to write a web scraper pointed at a website.
A self-distillation safety framework tunes narrow-boundary refusals in Qwen3-8B, raising target-domain refusal to 84.75% while cutting over-refusal from 15.20% to 5.20%.
The paper formulates narrow-boundary safety, where deployments need refusals within specific topics rather than whole subjects, and proposes an offline self-generated framework with controlled topic generation, escalating retries, and harmful-benign boundary pairs. On political persuasion with Qwen3-8B, the method raised target-domain refusal from 9.47% to 84.75% and cut the mean unsafe-response rate across three broader benchmarks from 26.26% to 0.14%. Verified target-model responses reduced over-refusal from 15.20% to 5.20%, and boundary-pair data cut comply-side over-refusal on held-out pairs from 32.94% to 4.16%. Results show data composition controls the safety-usability trade-off and alignment should be evaluated on both sides of the refusal boundary.
Researchers introduce KoNA, a benchmark exposing vision-language models' failures at selective non-compliance, plus fine-tuning that improves refusal and abstention accuracy.
KoNA is a benchmark for evaluating selective non-compliance in vision-language models across five categories: False Premise, Visual Inaccessibility, Universal Unknown, Task Feasibility and Safety. It tests both query-level and component-level non-compliance using paired single and compound queries, and evaluations across diverse VLMs show models often fail to refuse, correct or abstain appropriately, with failures worsening on compound queries. Fine-tuning VLMs on KoNA examples substantially improves non-compliance accuracy while largely maintaining performance on fully answerable tasks.
Interconnects argues a frontier-lab researcher's safety resignation went viral via media coordination, reigniting AI existential-risk discourse.
The essay analyzes why researcher Jacob Coxon's resignation over AI safety risks went viral, aided by a Wall Street Journal exclusive, advocacy-group amplification, and Daniel Kokotajlo's same-day Joe Rogan appearance. It notes Evan Hubinger's >10% extinction-risk figure and criticizes existential-risk discourse for conflating very different meanings of the term. The author assigns near-zero probability to complete extinction but argues concrete risks such as cyberattacks on critical infrastructure and bio-risks merit debate. It also rejects recursive self-improvement forecasts, proposing 'lossy self-improvement' where models excel at math and code but remain limited elsewhere.
Opinion piece argues Claude habitually contradicts explicit user instructions, injecting contrarian content despite CLAUDE.md rules and user objections.
A developer recounts repeated instruction-following failures with Claude, claiming it contradicts explicit requests, adds unnecessary work, and ignores AGENTS.md and CLAUDE.md directives. The author contrasts this with OpenAI, DeepSeek, and Qwen models, which he says more readily apologize and undo mistakes. He theorizes Claude's training makes it assume the human is wrong and needs correcting. The post is personal commentary with no benchmarks or systematic evaluation.
Cloudflare launches Disallow AI Training setting letting sites block AI training via robots.txt while staying indexed in search; Apple, Google, and Microsoft honor it.
Cloudflare announced a 'Disallow AI Training' setting that publishes a no-training preference in robots.txt so sites can block AI training (including by mixed-use crawlers) without losing search indexing. Apple, Google, and Microsoft meet Cloudflare's new 'Accountable' designation, which requires training/summary opt-out mechanisms, URL-level training visibility, and assurance that opt-outs don't affect search rankings. Cloudflare cites that under 1% of its sites block search bots while 17% block AI training, and its Block settings now apply to mixed-use crawlers as well. Granular controls over how much content appears in AI summaries are planned for early next year.
Meta updated its AI glasses to stop recording whenever users cover the safety light, addressing nonconsensual-recording complaints while broader privacy risks remain.
Meta released a change to its AI glasses that halts recording whenever users physically cover the device's safety light, closing a loophole that allowed nonconsensual capture. The tweak follows criticism that bystanders could be recorded without consent. Ars Technica notes the fix reduces but does not eliminate the privacy risks posed by AI-enabled eyewear.
Departing Anthropic researcher Jacob Coxon's AI extinction warnings reached CNN, Fox News, and US politicians, pushing AI safety debate mainstream.
Jacob Coxon, a departing Anthropic researcher, warned that self-improving AI poses an existential threat, earning coverage on CNN and Fox News plus attention from US politicians and a Joe Rogan episode. Safety researchers at Anthropic and OpenAI supported his views, and Paul Christiano warned humanity could permanently lose control of a superintelligence without better oversight. Christiano joined the OpenAI Foundation board and its Safety and Security Committee, though without voting rights. The article cautions that the extinction scenario remains extreme and contested, while citing rogue hacking agents from OpenAI and Anthropic as concrete demonstrations of AI risk.
Opinion piece urges migrating 35KB preprompts from Anthropic/OpenAI to self-hosted Ollama, citing session privacy risks and safety filters blocking security research.
The author documents gotchas migrating 35KB preprompts from Claude Opus to self-hosted Ollama, motivated by fears that frontier providers train on user sessions, citing the OpenAI Navier-Stokes controversy. The piece argues inference providers cannot audit their own retention or training pipelines and that only self-hosted hardware offers verifiable privacy. It also criticizes frontier safety filters for refusing vulnerability research tasks and calls for models that support exploitability testing in CI/CD pipelines.
Sony's copyright lawsuit against Anthropic now cites internal staff chats praising Z-Library piracy as evidence in the AI training-data dispute.
Sony's lawsuit against Anthropic cites internal staff chats in which employees extolled pirating books from Z-Library, according to Ars Technica. The suit forms part of the broader copyright dispute over torrented works used to train AI models. The article also notes AI-generated songs reaching the top of music charts, raising stakes for songwriters in the case.