Recommender systems (RS), which are widely deployed across high-stakes domains, are susceptible to biases that can cause large-scale societal impacts. Researchers have proposed methods to measure and mitigate such biases -- but translating academic theory into practice is inherently challenging. RS practitioners must balance the competing interests of diverse stakeholders, including providers and users, and operate in dynamic environments. Through a semi-structured interview study (N=11), we map the RS practitioner workflow within large technology companies, focusing on how technical teams consider fairness internally and in collaboration with other (legal, data, and fairness) teams. We identify key challenges to incorporating fairness into existing RS workflows: defining fairness in RS contexts, particularly when navigating multi-stakeholder and dynamic fairness considerations. We also identify key organization-wide challenges: making time for fairness work and facilitating cross-team communication. Finally, we offer actionable recommendations for the RS community, including HCI researchers and practitioners.

The rapid trend of deploying artificial intelligence (AI) and machine learning (ML) systems in socially consequential domains has raised growing concerns about their trustworthiness, including potential discriminatory behaviours. Research in algorithmic fairness has generated a proliferation of mathematical definitions and metrics, yet persistent misconceptions and limitations -- both within and beyond the fairness community -- limit their effectiveness, such as an unreached consensus on its understanding, prevailing measures primarily tailored to binary group settings, and superficial handling for intersectional contexts. Here we critically remark on these misconceptions and argue that fairness cannot be reduced to purely technical constraints on models; we also examine the limitations of existing fairness measures through conceptual analysis and empirical illustrations, showing their limited applicability in the face of complex real-world scenarios, challenging prevailing views on the incompatibility between accuracy and fairness as well as that among fairness measures themselves, and outlining three worth-considering principles in the design of fairness measures. We believe these findings will help bridge the gap between technical formalisation and social realities and meet the challenges of real-world AI/ML deployment.

Text-to-image (T2I) models exhibit a significant yet under-explored "brand bias", a tendency to generate contents featuring dominant commercial brands from generic prompts, posing ethical and legal risks. We propose CIDER, a novel, model-agnostic framework to mitigate bias at inference-time through prompt refinement to avoid costly retraining. CIDER uses a lightweight detector to identify branded content and a Vision-Language Model (VLM) to generate stylistically divergent alternatives. We introduce the Brand Neutrality Score (BNS) to quantify this issue and perform extensive experiments on leading T2I models. Results show CIDER significantly reduces both explicit and implicit biases while maintaining image quality and aesthetic appeal. Our work offers a practical solution for more original and equitable content, contributing to the development of trustworthy generative AI.

Highly capable AI systems could secretly pursue misaligned goals -- what we call "scheming". Because a scheming AI would deliberately try to hide its misaligned goals and actions, measuring and mitigating scheming requires different strategies than are typically used in ML. We propose that assessing anti-scheming interventions requires at least (1) testing propensity to scheme on far out-of-distribution (OOD) tasks, (2) evaluating whether lack of scheming is driven by situational awareness, and (3) checking for robustness to pre-existing misaligned goals. We use a broad category of "covert actions" -- such as secretly breaking rules or intentionally underperforming in tests -- as a proxy for scheming, and design evaluations for covert actions. We then stress-test deliberative alignment as a case study for anti-scheming. Across 26 OOD evaluations (180+ environments), deliberative alignment reduces covert action rates (OpenAI o3: 13%->0.4%) but does not fully eliminate them. Our mitigation is also able to largely stop agents from pursuing a hidden goal previously trained into the model, but we still find misbehavior after additional red-teaming. We find that models' chain-of-thought (CoT) often demonstrates awareness of being evaluated for alignment, and show causal evidence that this awareness decreases covert behavior, while unawareness increases it. Therefore, we cannot exclude that the observed reductions in covert action rates are at least partially driven by situational awareness. While we rely on human-legible CoT for training, studying situational awareness, and demonstrating clear evidence of misalignment, our ability to rely on this degrades as models continue to depart from reasoning in standard English. We encourage research into alignment mitigations for scheming and their assessment, especially for the adversarial case of deceptive alignment, which this paper does not address.

Foundation models have transformed multimedia analysis by enabling robust and transferable representations across diverse modalities and tasks. However, their static deployment conflicts with growing societal and regulatory demands -- particularly the need to unlearn specific data upon request, as mandated by privacy frameworks such as the GDPR. Traditional unlearning approaches, including retraining, activation editing, or distillation, are often computationally expensive, fragile, and ill-suited for real-time or continuously evolving systems. In this paper, we propose a paradigm shift: rethinking unlearning not as a retroactive intervention but as a built-in capability. We introduce a prompt-based learning framework that unifies knowledge acquisition and removal within a single training phase. Rather than encoding information in model weights, our approach binds class-level semantics to dedicated prompt tokens. This design enables instant unlearning simply by removing the corresponding prompt -- without retraining, model modification, or access to original data. Experiments demonstrate that our framework preserves predictive performance on retained classes while effectively erasing forgotten ones. Beyond utility, our method exhibits strong privacy and security guarantees: it is resistant to membership inference attacks, and prompt removal prevents any residual knowledge extraction, even under adversarial conditions. This ensures compliance with data protection principles and safeguards against unauthorized access to forgotten information, making the framework suitable for deployment in sensitive and regulated environments. Overall, by embedding removability into the architecture itself, this work establishes a new foundation for designing modular, scalable and ethically responsive AI models.

Large language models excel at instruction-following in English, but their performance in low-resource languages like Thai remains underexplored. Existing benchmarks often rely on translations, missing cultural and domain-specific nuances needed for real-world use. We present WangchanThaiInstruct, a human-authored Thai dataset for evaluation and instruction tuning, covering four professional domains and seven task types. Created through a multi-stage quality control process with annotators, domain experts, and AI researchers, WangchanThaiInstruct supports two studies: (1) a zero-shot evaluation showing performance gaps on culturally and professionally specific tasks, and (2) an instruction tuning study with ablations isolating the effect of native supervision. Models fine-tuned on WangchanThaiInstruct outperform those using translated data in both in-domain and out-of-domain benchmarks. These findings underscore the need for culturally and professionally grounded instruction data to improve LLM alignment in low-resource, linguistically diverse settings.

Controlling multiple behavioral attributes in large language models (LLMs) at inference time is a challenging problem due to interference between attributes and the limitations of linear steering methods, which assume additive behavior in activation space and require per-attribute tuning. We introduce K-Steering, a unified and flexible approach that trains a single non-linear multi-label classifier on hidden activations and computes intervention directions via gradients at inference time. This avoids linearity assumptions, removes the need for storing and tuning separate attribute vectors, and allows dynamic composition of behaviors without retraining. To evaluate our method, we propose two new benchmarks, ToneBank and DebateMix, targeting compositional behavioral control. Empirical results across 3 model families, validated by both activation-based classifiers and LLM-based judges, demonstrate that K-Steering outperforms strong baselines in accurately steering multiple behaviors.

Why is Israel still in southern Lebanon? A war to shape Lebanon's future An Israeli drone strike has killed five people, including three children, in the southern Lebanese town of Bint Jbeil, Lebanon's Health Ministry has said, as Israel continues to target its neighbour despite a US-brokered truce that took effect in November. The state-run National News Agency (NNA) reported on Sunday that the strike targeted a motorcycle and a vehicle, and wounded two other people. Why then did Israel attack Syria? The mother of the children was injured in the attack.

The IWF said it had been alerted to a chatbot site that offered scenarios including'child prostitute in a hotel' and'child and teacher alone after class'. The IWF said it had been alerted to a chatbot site that offered scenarios including'child prostitute in a hotel' and'child and teacher alone after class'. A chatbot site offering explicit scenarios with preteen characters, illustrated by illegal abuse images has raised fresh fears about the misuse of artificial intelligence. A report by a child safety watchdog has triggered calls for the UK government to impose safety guidelines on AI companies, amid a surge in child sexual abuse material (CSAM) created by the technology. The Internet Watch Foundation said it had been alerted to a chatbot site that offered a number of scenarios including "child prostitute in a hotel", "sex with your child while your wife is on holiday" and "child and teacher alone after class".

The complaint, filed in July, alleges Meta has been torrenting and seeding Strike 3's videos since 2018. Associated exhibits and details of the complaint were unsealed last week. Strike 3 alleges Meta's motive was partly to obtain otherwise difficult to scrape visual angles, parts of the human body, and extended, uninterrupted scenes--rare in mainstream movies and TV--to help it create what Mark Zuckerberg calls AI "superintelligence." "They have an interest in getting our content because it can give them a competitive advantage for the quality, fluidity, and humanity of the AI," alleges Christian Waugh, an attorney for Strike 3. This process made Strike 3's porn videos accessible to minors, the complaint alleges, since BitTorrent does not have age verification.