Large language models (LLMs) have achieved impressive results across a range of natural language processing tasks, but their potential to generate harmful content has raised serious safety concerns. Current toxicity detectors primarily rely on single-label benchmarks, which cannot adequately capture the inherently ambiguous and multi-dimensional nature of real-world toxic prompts. This limitation results in biased evaluations, including missed toxic detections and false positives, undermining the reliability of existing detectors. Additionally, gathering comprehensive multi-label annotations across fine-grained toxicity categories is prohibitively costly, further hindering effective evaluation and development. To tackle these issues, we introduce three novel multi-label benchmarks for toxicity detection: \textbf{Q-A-MLL}, \textbf{R-A-MLL}, and \textbf{H-X-MLL}, derived from public toxicity datasets and annotated according to a detailed 15-category taxonomy. We further provide a theoretical proof that, on our released datasets, training with pseudo-labels yields better performance than directly learning from single-label supervision. In addition, we develop a pseudo-label-based toxicity detection method. Extensive experimental results show that our approach significantly surpasses advanced baselines, including GPT-4o and DeepSeek, thus enabling more accurate and reliable evaluation of multi-label toxicity in LLM-generated content.

With the growth of digital financial systems, robust security and privacy have become a concern for financial institutions. Even though traditional machine learning models have shown to be effective in fraud detections, they often compromise user data by requiring centralized access to sensitive information. In IoT-enabled financial endpoints such as ATMs and POS Systems that regularly produce sensitive data that is sent over the network. Federated Learning (FL) offers a privacy-preserving, decentralized model training across institutions without sharing raw data. FL enables cross-silo collaboration among banks while also using cross-device learning on IoT endpoints. This survey explores the role of FL in enhancing financial security and introduces a novel classification of its applications based on regulatory and compliance exposure levels ranging from low-exposure tasks such as collaborative portfolio optimization to high-exposure tasks like real-time fraud detection. Unlike prior surveys, this work reviews FL's practical use within financial systems, discussing its regulatory compliance and recent successes in fraud prevention and blockchain-integrated frameworks. However, FL deployment in finance is not without challenges. Data heterogeneity, adversarial attacks, and regulatory compliance make implementation far from easy. This survey reviews current defense mechanisms and discusses future directions, including blockchain integration, differential privacy, secure multi-party computation, and quantum-secure frameworks. Ultimately, this work aims to be a resource for researchers exploring FL's potential to advance secure, privacy-compliant financial systems.

Large Language Models are susceptible to jailbreak attacks that bypass built-in safety guardrails (e.g., by tricking the model with adversarial prompts). We propose Concept Alignment and Concept Manipulation CALM, an inference-time method that suppresses harmful concepts by modifying latent representations of the last layer of the model, without retraining. Leveraging concept whitening technique from Computer Vision combined with orthogonal projection, CALM removes unwanted latent directions associated with harmful content while preserving model performance. Experiments show that CALM reduces harmful outputs and outperforms baseline methods in most metrics, offering a lightweight approach to AI safety with no additional training data or model fine-tuning, while incurring only a small computational overhead at inference.

State's Republican-led public service commission to decide on power expansion and prices, as Democrats vie for voice Georgia is facing the largest demand for electricity in its history, driven by nation-leading datacenter construction. The Georgia Power company has made an unprecedented bid to the agency that oversees the utility for about 10 additional gigawatts of energy in the coming years - enough to power 8.3m homes, at an estimated cost of nearly $16bn, according to the Southern Environmental Law Center . But those huge numbers are not primarily for homes or local businesses in Georgia . Instead about 80% of the company's ask is driven by datacenters, primarily for artificial intelligence, according to Tom Krause, spokesperson for the state's public service commission, or PSC. It is the largest increase ever considered by the commission in a multiyear plan and comes as the Atlanta metro area led the nation in datacenter construction last year - a phenomenon playing out across the US and increasingly sparking protests and pushback.

OpenAI patched the ShadowLeak attack in August 2025 after hackers exploited ChatGPT to exfiltrate Gmail data through invisible email prompts without user interaction.

An illustration of how AI manipulates and changes images. An illustration of how AI manipulates and changes images. Ask Google's AI video tool to create a film of a time-travelling doctor who flies around in a blue British phone booth and the result, unsurprisingly, resembles Doctor Who . And if you ask OpenAI's technology to do the same, a similar thing happens. What's wrong with that, you may think?

Federal regulators investigate Tesla's Full Self-Driving technology over safety concerns, including 14 crashes and 23 injuries from traffic violations.

Plus: A secret FBI anti-ransomware task force gets exposed, the mystery of the CIA's Kryptos sculpture is finally solved, North Koreans busted hiding malware in the Ethereum blockchain, and more. In a stunning new study, researchers at UC San Diego and the University of Maryland revealed this week that satellites are leaking a wealth of sensitive data completely unencrypted, from calls and text messages on T-Mobile to in-flight Wi-Fi browsing sessions, to military and police communications. And they did this with just $800 in off-the-shelf equipment. Face recognition systems are seemingly everywhere. But what happens when this surveillance and identification technology doesn't recognize your face as a face?

Things to Do in L.A. Tap to enable a layout that focuses on the article. State Sen. Scott Wiener stands in front of a mural at Oasis, a drag show he helped the owners launch in San Francisco. He intends to run for Nancy Pelosi's long-held congressional seat. The former House speaker has not said whether she will seek another term. This is read by an automated voice.

Technology Engineering OceanGate's'Titan' went on 7 dives with a damaged hull before implosion Investigators found that the submersible's exterior featured'multiple anomalies' as early as 2022. Breakthroughs, discoveries, and DIY tips sent every weekday. The United States National Transportation Safety Board (NTSB) recently concluded its investigation into the OceanGate submersible disaster . According to the summary report released on October 15, an already weakened hull caused the deep sea tourist vessel to implode while it was en route to visit the wreckage of the RMS in June 2023, killing all five passengers on board. But according to their findings, investigators noted that the submersible wasn't damaged shortly before its final voyage.