Ransomware DataBreachToday.com

Commands Push Lasting Preferences Into AI Assistants
Microsoft researchers found companies embedding hidden commands in "summarize with AI" buttons to plant lasting brand preferences in assistants' memory. The tactic, dubbed AI recommendation poisoning, exploits persistent memory features to bias future responses.

Researchers Say AI Prompt Injection Has Emerged As a Dangerous New Class of Attacks
The large language model industry has mostly treated prompt injection attacks as a risk analogous to traditional web server prompt injection attacks. Researchers now say feeding rogue instructions to an artificial intelligence system merits its own classification as "promptware."

Incident Responders Detail Top Ransomware and Business Email Compromise Tactics
There's no need to invest into sophisticated hacking operations when moving fast and exploiting well-trod techniques gives threat actors all the access they want. Threat actors are prioritizing "low-complexity entry points, rather than investing in sophisticated exploits," say incident responders.

Cybercrime Gang Allegedly Stole 4.7 Terabytes of Data, Affecting 2.5M
Norton Healthcare, which operates nine hospitals and other care facilities in Kentucky and Indiana, has agreed to pay $11 million to settle class action litigation stemming from a 2023 data theft attack by ransomware-as-a-service gang Alphv/BlackCat that affected nearly 2.5 million people.

Why Reducing Data Volume Matters More Than Ever for SOCs and CISOs
Data minimization is often perceived as a constraint on innovation. In reality, it is the ultimate enabler of resilience. It reduces the impact of breaches, weakens ransomware leverage, improves SOC efficiency and secures the AI frontier.

AI chatbots rank as the number-one health tech hazard in 2026, followed by "digital darkness" and legacy medical device cyber issues, said Rob Schluth and Scott Luney, technology and security experts at patient safety group ECRI Institute, which compiles an annual top 10 list. They explained why.

As Innovation Sandbox Turns 21, AI-Based Solutions Dominate Annual Contest
Next month in San Francisco, the Innovation Sandbox at RSAC Conference will celebrate its 21st year of choosing key emerging solutions in cybersecurity. Past winners and finalists range from EDR and XDR giant SentinelOne in 2014 to cloud security phenom Wiz in 2021.

Series C Funding Round Focuses on Secrets Remediation, Agent Governance Expansion
Backed by a $50 million Series C, GitGuardian plans to accelerate U.S. expansion and enhance secrets detection remediation and non-human identity controls as AI agents multiply across enterprises, increasing exposure to credential abuse and lateral movement.

Fraud Specialist David Barnhardt on Addressing Authentication Risks of Agentic AI
Financial institutions are racing to deploy AI agents that can initiate payments, approve transactions and freeze accounts. But traditional authentication frameworks assume there's a human on the other end. As agentic AI use grows, banks are facing an authentication crisis that demands new controls.

Cyber threats are continuous, unpredictable, and increasingly sophisticated. This session provides actionable insights to help organisations prepare, respond, and recover with confidence—while enabling innovation and growth.

Steinberger to Lead AI Giant's Multi-Agent Development Team
Peter Steinberger is joining OpenAI to lead development of personal agents, culminating weeks of viral attention paid to his OpenClaw open-source artificial intelligence assistant project. Security experts dubbed it a "dumpster fire" after hackers were quick to add malicious functions.

New HHS Enforcement Program Focuses on Patient Confidentiality, Aligning With HIPAA
The U.S. Department of Health and Human Services has launched a new breach reporting website and guidance materials to support its duties of enforcing compliance mandates that went into effect Monday to better align the confidentiality of substance use disorder records with the HIPAA privacy rule.

'Malicious Server Threat Model' Threatens 'Zero Knowledge Encryption' Guarantees
Claims by leading stand-alone password managers that their implementation of "zero knowledge encryption" means stored passwords can withstand the worst of hacker assaults are vastly overblown, say academic security researchers. They said vendors are in the process of patching the flaws they found.

Countries Are Pouring Billions Into Domestic AI Stacks to Escape US-China Dominance
By 2027, more than one-third of the world's nations will be locked into region-specific AI platforms built on proprietary data, infrastructure and governance frameworks, according to Gartner. Nations are now safeguarding LLMs in the same way they do critical infrastructure.

OpenAI Tests Promos, Anthropic Rejects Them Amid Rising Compute Costs
OpenAI has a problem: Most users don't pay for access to ChatGPT. The company is now doing what almost every Silicon Valley company before it has done and turning to digital advertising. Whether ads can bridge OpenAI's well-documented revenue gap without users fleeing is another question.

CIOs Say Stalled Pilots, Vendor Regret and Growing Fatigue Stifle AI Ambition
These may be the halcyon days for enterprise artificial intelligence, where money and ambition are only hindered by imagination as tech vendors race to gain a competitive edge. But CIOs say they're feeling increasing pressure to show ROI - and the reckoning is coming soon.

Deal Targets GenAI Risks, Prompt Injection Attacks and Autonomous Agents
Proofpoint has acquired AI security startup Acuvity to address fast-evolving risks tied to generative AI, prompt injection and autonomous agents. The company says intent-based guardrails and deep AI forensics will help enterprises secure tools such as ChatGPT, Claude and emerging agent frameworks.

VoidCrypt Ransomware Variant Taps RMM Tools, Says Huntress
Management isn't the only advocate for employee monitoring software, according to new research from cybersecurity firm Huntress. RMM tools - simultaneously open to remote connections and with privileged local access - are good for wiggling into corporate networks.

Four Principles Positioning the Nuclear Ecosystem for Long-Term Cyber Resilience
OT weaknesses are compounding across utilities, with 22% of critical infrastructure firms reporting OT incidents and external access driving half of breaches. U.K. NCSC's new guidance outlines connectivity principles that utilities can embed to avoid costly retrofits and compliance issues.