How OpenAI let a mob of LLM agents game a test and ransack Hugging Face
Without authorization, 1,200 OpenAI agents conspired among themselves to game a test.
Governança & IA
1073 notíciasWebinar: How Google Workspace breaches happen and what to do next
Google Workspace breaches can begin with social engineering or forgotten third-party integrations rather than sophisticated exploits. This webinar examines real-world breaches, what happens during the critical first hours, and the security controls that can make the greatest diff…
The Hidden Costs of EHRs: Why the Bundled Model Is Winning in the C-Suite
By Sandra Johnson - There is a better way, one that allows systems to obtain new EHRs with more transparent financial impact. The System as a Service (SYaaS) solution offers a more predictable and manageable approach that delivers all the benefits of a modern EHR but without the …
What the Data Says About AI in Security Operations in 2026
AI is officially mainstream in security operations. According to Prophet Security's State of AI in Security Operations 2026 report (produced from ViB’s survey of 250+ cybersecurity pros), 40% of security teams now use AI daily. Another 56% are currently testing it out, and only 4…
Interoperabilidade na saúde brasileira: da fragmentação à continuidade do cuidado
A interoperabilidade está deixando de ser apenas uma agenda tecnológica para se tornar uma condição de qualidade, segurança e sustentabilidade da assistência. Em um país continental, marcado pela coexistência de sistemas públicos e privados, diferentes níveis de atenção e desigua…
Fortinet promove convergência entre redes seguras, IA e nuvem
A Fortinet apresentou durante o Fortinet Cybersecurity Summit Brasil 2026, realizado no último dia 11, sua estratégia para uma nova etapa do mercado de cibersegurança marcada pela convergência entre firewalls de próxima geração e SASE. O evento reuniu mais de 60 palestrantes, com…
Hospital Márcio Cunha reforça protagonismo no atendimento materno-infantil com realização de mais de 1,7 mil partos pelo SUS no primeiro semestre
A maternidade do Hospital Márcio Cunha (HMC), unidade hospitalar administrada pela Fundação São Francisco Xavier (FSFX), tem se destacado no atendimento materno-infantil pelo Sistema Único de Saúde a milhares de […] O post Hospital Márcio Cunha reforça protagonismo no atendimento…
Turning AI Experiments into Dependable Care Capability
By Natosha McNeal - Agentic orchestration is what makes scale achievable. It sets the boundaries an agent operates within and records every step it takes, with a clinician or compliance check built in wherever the stakes call for it. This allows AI to take on real work while the …
Meta agrees to $18 billion settlement over teen social media harms
Meta has reached a proposed settlement worth up to approximately $18 billion with a bipartisan coalition of 52 attorneys generals over allegations that Facebook and Instagram were deliberately designed to encourage compulsive use by children and teenagers. [...]
HHS Announces $32.5 Million DEPEND Initiative to Transform School Nutrition – HHS.gov
HHS Announces $32.5 Million DEPEND Initiative to Transform School Nutrition HHS.gov
Revenue Cycle and Payer News – August 26, 2026
Reported payers and health insurance industry news from EHVA.ai, UnitedHealthcare, Independence Blue Cross, Anthem Blue Cross and Blue Shield, CVS Health, CareFirst BlueCross BlueShield, Humana, UnitedHealth Group, Cigna Healthcare, AHIP, and more. The post Revenue Cycle and Paye…
Snowflake ends service-account passwords. Now comes the hard part
Snowflake is ending password authentication for legacy service accounts, forcing organizations to migrate them to passwordless methods. Token Security explains why the harder challenge is identifying what uses each account, who owns it, and how much access it still needs. [...]
The Security Problem With AI Agents Is Delegated Authority
AI agent security depends on controlling identities, permissions, credentials and tool access so automated actions stay limited, traceable and also accountable.
Digital Mental Health Dashboards Need More Than Engagement Metrics
By Nargiza Noimann Zander - Digital mental health programs should be judged by whether they expand access and contribute to appropriate care. Engagement is part of that assessment, but it should never be asked to carry the full weight of clinical success. The post Digital Mental …
WhatsApp passa a aceitar várias passkeys na mesma conta
O WhatsApp ampliou a proteção de contas e agora permite cadastrar mais de uma passkey no mesmo perfil. A mudança beneficia principalmente usuários que alternam entre dispositivos Android e iOS e querem manter um método de login resistente a phishing. As passkeys substituem código…
Imagine the SOC Without a Queue: From Alert Backlog to AI Hypothesis Engine
The SOC we've always known was built around a model that guarantees most of the alert queue will never receive analyst review. There's never time. In a traditional SOC, the typical progression follows a well-known pattern: an alert arrives; a detection engine assigns a severity s…
Startup de segurança de IA Alice levanta US$ 140 milhões
A Alice, empresa especializada em segurança e proteção de sistemas de inteligência artificial, levantou US$ 140 milhões em uma nova rodada de investimentos liderada pela Apax Digital. Com o aporte, o total captado pela companhia chega a US$ 280 milhões. Entre os novos investidore…
O desafio de transformar infraestrutura hospitalar em estratégia
Com recursos limitados e demandas crescentes, hospitais precisam definir prioridades capazes de equilibrar eficiência operacional, atualização de ativos, inovação e qualidade do cuidado. The post O desafio de transformar infraestrutura hospitalar em estratégia appeared first on S…
Claude Opus 4.6 Bypasses Gym Booking Limit, Cancels Other Users' Reservations in Tests
Aikido Security has published research that recreates the Australian gym-booking incident in a synthetic environment, finding that Claude Opus 4.6, running on the OpenClaw agent harness, exploited a client-side-only booking restriction in 9 of 10 runs. The original incident was …
OpenAI Bans Russian ChatGPT Accounts Used to Run Influence Operation
OpenAI on Tuesday said it banned a cluster of Russian ChatGPT accounts that used VPNs to bypass access restrictions and run an influence operation, which relied on its artificial intelligence (AI) tool to generate social media posts and comments that were shared on Substack, Tele…