Stay up to date on MCP vulnerabilities, AI-driven cyberattacks, prompt injection threats, and the security developments that matter to AI practitioners. Search and filter instantly — right in your browser.
200 articles
Try a different search term or clear the topic filter.
Learn how OX VibeSec protects against the Anthropic MCP vulnerability with layered defenses that stop insecure AI-generated code,...
Companies are falling further behind in their ability to defend against the cybersecurity threats they're most concerned about.
Most of the conversation around AI in cybersecurity focuses on how attacks are getting faster and more sophisticated.
The chief of Europe's securities regulator said the risks from and potential speed of cyberattacks are growing, joining a chorus of...
Anthropic's new AI model, Claude Mythos, warns of a heightened threat to elections — but not a new one. Derek Tisler Photo Derek Tisler.
A lack of guardrails in internal processes is making governments more vulnerable, a cybersecurity expert says.
We see a lot of doom and gloom about the potential negative impacts of artificial intelligence, particularly centered on how it could create...
The World Economic Forum (WEF) warns that the emergence of advanced AI systems such as Anthropic's Mythos marks a turning point for...
S. Korea's intelligence agency has warned that AI capable of independently carrying out cyberattacks is emerging as a potential cyber...
South Korea's National Intelligence Service issued a government-wide advisory warning that next-generation AI models can autonomously...
Infosecurity Magazine reports that the lack of strategy on the use and governance of AI agents is increasing organizations' risk of...
With great power comes great responsibility. Before artificial intelligence became readily accessible, cybersecurity risks were...
Using generative AI to design, train, or perform steps within a machine-learning system is risky, argues computer scientist Micheal Lones in...
A design choice in the MCP SDKs allows remote code execution across the AI supply chain.
Dutch intelligence warns Russia is using AI to accelerate cyberattacks on Europe, posing a growing threat to critical infrastructure and...
The boom in artificial intelligence (AI) and geopolitical tensions are giving rise to increased cybersecurity threats, according to the head...
The launch of Claude Mythos is a signal, not the threat itself. Multiple frontier AI models already enable sophisticated cyberattacks,...
Australia and New Zealand's central banks are monitoring developments around Anthropic PBC's new Mythos AI model, which the company says is...
Anthropic has said Mythos can identify and exploit vulnerabilities 'in every major operating system and every major web browser when...
AI-powered integrated developer environments (IDEs) like Cursor, VS Code, and Windsurf now include agents that utilize Model Context...
The Cybersecurity and Infrastructure Security Agency doesn't have access to Anthropic's powerful new Mythos Preview model, even though some...
Russia is deploying artificial intelligence to accelerate its cyberattacks on Europe, Dutch military intelligence warned Tuesday — and the...
Cyberattacks on businesses are no longer rare incidents—they've become an almost routine threat in today's digital landscape.
A critical flaw in Anthropic's Model Context Protocol (MCP) exposes over 150 million downloads to potential compromise. The vulnerability...
On Friday, Barclays PLC CEO CS Venkatakrishnan warned that a powerful new AI model from Anthropic could significantly heighten cyber risks...
A critical vulnerability in Anthropic's Model Context Protocol (MCP) is putting millions of systems at risk of full compromise, according to...
An architectural vulnerability baked into the core of Anthropic's Model Context Protocol (MCP) exposes millions of AI applications to remote...
Banks are reportedly increasing their efforts to defend themselves against cyberattacks amid new technological breakthroughs.
Banks are stepping up cybersecurity as **AI** materially changes the threat landscape. Research from **Kroll** shows **76%** of companies...
A critical vulnerability discovered in Anthropic's MCP has exposed over 150 million downloads and up to 200000 servers to complete takeover.
Compromised Context.ai integration let attackers inherit Vercel employee access and reach internal systems, exposing a limited set of...
MCP design flaw enables RCE across 7000+ servers and 150M downloads, impacting AI SDKs and supply chains.
A critical vulnerability in Flowise and multiple AI frameworks has been discovered by OX Security, exposing millions of users to remote code...
Anthropic CEO Dario Amodei met White House Chief of Staff Susie Wiles as the government reacts to its Mythos system that automates some of...
Barclays CEO warned that Anthropic's powerful Mythos AI could expose and exploit vulnerabilities in global banking systems.
We see a lot of doom and gloom about the potential negative impacts of artificial intelligence, particularly centered on how it could create...
A newly disclosed critical vulnerability in Flowise, linked to Anthropic's Model Context Protocol (MCP), is raising serious concerns across the AI and...
On April 6, cancer patients at Brockton Hospital in Massachusetts showed up for chemotherapy infusions and were told to go home.
A vulnerability in Anthropic's open source Model Context Protocol is putting up to 200000 servers at risk of complete takeover.
Breaches, nation-state attacks, AI-powered threats, among other things, bring these 4 cybersecurity companies in focus. Take a look.
A design flaw – or expected behavior based on a bad design choice, depending on who is telling the story – baked into Anthropic's official...
Unsafe defaults in MCP configs open servers to possible remote code execution, as evidenced by several commercial services and open-source...
'GPT‑5.4‑Cyber' is specifically made to be better at breaking through security, and with fewer restrictions.
Security researchers have warned of a “critical, systemic” vulnerability in the model context protocol (MCP) which could have a significant...
Artificial intelligence (AI) technology offers a positive contribution to improving our cybersecurity. However, it can also facilitate and...
The cybersecurity company OX Security reported that the MCP protocol of Anthropic has a design flaw, and its SDK's STDIO interface could...
Attackers can abuse the near-maximum severity flaw in nginx-ui to restart, create, modify, and delete NGINX configuration files.
Advances in AI mean the window of time between published IT flaws and their exploitation has effectively vanished, says Parmy Olson for...
The question for executives is no longer whether artificial intelligence will affect cybersecurity. It is whether their organization is...
Project Glasswing will attempt to use Claude Mythos's capability to identify vulnerabilities for defensive purposes.
A critical Nginx UI vulnerability that allows attackers to take full control of servers has been exploited in the wild.
A critical Nginx UI vulnerability that allows attackers to take full control of servers has been exploited in the wild.
A 'by design' flaw in Anthropic's Model Context Protocol (MCP) could allow silent command execution and full system compromise,...
CVE-2026-33032 exposes nginx-ui to unauthenticated takeover via MCP endpoint, impacting 2600+ instances with active exploitation.
OX Security's full disclosure advisory for command injection vulnerabilities in MCP STDIO configurations, covering 10 CVEs.
OX researchers uncovered an architectural RCE vulnerability in Anthropic's Model Context Protocol — affecting 150M+ downloads.
Built by a veteran security team and led by a former Google and Mandiant executive, Mallory delivers intelligence that drives action for...
AI toolchains and compromised dependencies are turning stealth attacks into a systemic risk across enterprise infrastructure.
Jamie Dimon said AI is a double-edged sword: “it's made it worse, it's made it harder,” creating new cyber vulnerabilities even as it may...
CISOs must prepare for AI-powered cyberattacks as Anthropic's Mythos speeds up vulnerability discovery and exploitation, creating an “AI...
Anthropic's Mythos, a new AI model the company and cybersecurity experts warn could supercharge complex cyberattacks, poses significant...
Anthropic's newest AI model, Claude Mythos Preview, has sent the banking world into high alert. Announced on April 7, the model is so...
Over a quarter (26%) of UK business leaders are not using AI to enhance security; More than half of businesses have been victims of a...
First, tech giant Microsoft ($MSFT) was frantically trying to shoehorn artificial intelligence (AI) into just about everything it made.
Federal officials warn banks about AI-driven cyberattacks, as AI tools make attacks more sophisticated and harder to stop.
Hackers are using generative AI to write phishing emails, build malware and scale cyberattacks faster than ever.
The AI Security Institute (AISI) conducted evaluations of Anthropic's Claude Mythos Preview (announced on 7th April) to assess its...
AI is accelerating cyberattacks, collapsing timelines and exposing new identity risks. Okta's Brett Winterford explains how attackers are...
Better AI means that scammers have more tools at their disposal to commit cybercrimes.
We see a lot of doom and gloom about the potential negative impacts of artificial intelligence, particularly centered on how it could create...
In an unusual move, the Treasury secretary and the Federal Reserve chair gathered bank executives to caution about cyberthreats posed by...
The rapid advancement in artificial intelligence across sectors is challenging both regulators and technology companies as they work to...
Anthropic's Claude Mythos Preview showed advanced ability to find and exploit vulnerabilities across computer systems. This is raising fears...
The steady advance of artificial intelligence models continues to raise serious concerns about the cybersecurity threats the technology...
Editor's note: A previous version of this article stated that this was Crowdstrike's first-ever positive GAAP net income,...
Editor's note: A previous version of this article stated that this was Crowdstrike's first-ever positive GAAP net income,...
AI is accelerating cyberattacks, collapsing timelines and exposing new identity risks. Okta's Brett Winterford explains how attackers are...
The design flaw in Flowise's Custom MCP node has allowed attackers to execute arbitrary JavaScript through unvalidated configurations.
CNBC's Andrew Ross Sorkin reports on the latest news.
Consider this the scary phase of artificial intelligence.
Project Glasswing is Anthropic's new AI model for software security.
A new initiative to secure the world's most critical software and give defenders a durable advantage in the coming AI-driven era of...
The company says it found strong cyberdefense skills in its forthcoming 'Mythos' model, which it will lend to the industry research effort.
We see a lot of doom and gloom about the potential negative impacts of artificial intelligence, particularly centered on how it could create...
Private AI developer Anthropic (ANTH.PVT) is partnering with an array of Big Tech names for its Project Butterfly cybersecurity initiative...
Apple, Google, and almost all of the rest of US big tech have signed up to Anthropic's Project Glasswing, that will use AI to improve the...
Anthropic, the artificial intelligence company that recently fought the Pentagon over the use of its technology, has built a new A.I. model...
Anthropic announced Claude Mythos Preview, which it said is an advanced AI model that excels at identifying weaknesses and security flaws...
The next generation of artificial intelligence-driven cyberattacks could far surpass anything previously encountered, according to warnings...
CVE-2025-59528 exploited in Flowise for over six months across 12000+ exposed instances, enabling full system compromise.
Anthropic, a prominent artificial intelligence company, has alleged that state-sponsored Chinese hackers used its own AI technology in an...
The crypto industry faces growing cybersecurity risks as AI tools lower the cost and skill needed to exploit software flaws, with over $1.4...
Anthropic said late last year that state-sponsored Chinese hackers had used its artificial intelligence technology in an effort to...
AI is accelerating cyberattacks, and most leaders aren't ready. Hise Gibson explains why traditional risk prevention strategies aren't...
Cyberattacks and data breaches have become a defining risk for modern businesses, particularly those operating in data-intensive sectors.
Cyberattacks are rising as hackers target small businesses and nonprofits using AI-driven scams, ransomware and email fraud.
A new EY report suggests AI will become a major force behind cybercrime in 2026 as programs help attackers hone their scamming skills.
A leaked blog post from AI company Anthropic said the next round of AI-driven cyberattacks could be far more difficult to stop.
Armis finds AI-driven cyberattacks are accelerating, widening the gap between perceived readiness and real-world defenses.
Leaked Anthropic notes and expert warnings suggest agentic AI models like Mythos could enable faster, larger cyberattacks.
The next wave of AI-powered cybersecurity attacks will be like nothing we've seen before.
Leaked Anthropic post warns of faster AI cyberattacks ... AI company Anthropic warned that the next wave of AI would pose a new type of...
Generative AI is upgrading cyberattacks, from 450% higher phishing click‑through rates to industrialized MFA bypass. Learn more.
As AI adoption accelerates, security leaders are being forced to rethink how they approach cyber risk. Traditional models built for static...
As AI adoption accelerates, security leaders are being forced to rethink how they approach cyber risk. Traditional models built for static...
Fortinet's Aamir Lakhani explains how AI is accelerating cyberattacks and forcing defenders to operate at machine speed.
Hostile actors, including state-linked entities, are leveraging artificial intelligence tools such as ChatGPT and other platforms to conduct...
AI will make cyberattacks faster and scalable as soon as this year, pushing companies in Latin America to rethink security, says Google...
AI is accelerating cyberattacks, and most leaders aren't ready. Hise Gibson explains why traditional risk prevention strategies aren't...
Iran's Revolutionary Guard names major US tech firms on April 1 attack list.
Cybersecurity professionals said Tuesday companies should act as though AI-enabled cyberattacks are looming, and try to reduce the “blast...
Cybersecurity professionals said Tuesday companies should act as though AI-enabled cyberattacks are looming, and try to reduce the “blast...
In a conversation with Dark Reading's Terry Sweeney, Gigamon's Gareth Maclachlan explains how AI is reshaping attacker economics.
ThreatLocker CEO Danny Jenkins explains why and how basic security controls can prevent up to 96% of breaches.
Artificial intelligence companies including OpenAI and Anthropic could, within six months, roll out advanced models capable of finding...
Object First, the ransomware-proof on-premises backup storage company, released the results of its new World Backup Day survey that examines...
DENVER, March 31, 2026--Object First, the ransomware-proof on-premises backup storage company, released the results of its new World Backup...
Cyberattacks now target human psychology at scale, reshaping how organizations defend and handle trust, said Sarah Gosler,...
A March 2026 report on AI cyber threats has found that Uruguay is the world's most resilient country against hacking attacks.
Record-breaking cyber attacks, undetectable malware and deepfakes that are indistinguishable from loved ones. Anthony Cuthbertson looks at...
AI-powered cyber threats are reshaping the global risk landscape, exposing critical vulnerabilities in emerging digital economies,
OpenAI and Anthropic are preparing to release next-generation systems capable of hacking complex systems easily, Axios has reported.
Top AI and government officials tell Axios CEO Jim VandeHei that Anthropic, OpenAI and other tech giants will soon release new models that...
Since the war began last month, hackers supporting Iran have launched thousands of cyberattacks on companies and organizations in both the...
AI has compressed cyberattack timelines from years to days, enabling rapid exploitation and automation. Robert T. Lee, chief AI officer and...
From industrialized zero-day exploits to artificial intelligence-driven operations in critical infrastructure, cybersecurity leaders warn...
The data and AI provider's launch of Lakewatch comes amid rising concerns over cyberattacks fueled by agentic AI.
Insurers are facing worsening aggregation exposure as threats scale across sectors, say TMHCC specialists.
Insurers are facing worsening aggregation exposure as threats scale across sectors, say TMHCC specialists.
Munich Re expects agentic AI to affect the frequency of cyberattacks more than severity in the near term, while new data shows ransomware...
N-able rolls out AI-led SOC detections to flag PowerShell abuse, suspicious DNS traffic and unusual Windows processes across multiple...
Experts said companies rushing to buy AI services risked letting their existing, still-vital defensive measures deteriorate.
By combining Zero Trust's "never trust, always verify" principles with AI-driven proactive protection, you can transform data recovery from...
Read the latest Microsoft Secure Access report for insights into why a unified identity and access strategy offers strong modern protection.
AI cyberattacks are rising fast, exposing limits of software security. Here is why institutions are shifting to hardware and what Ledger's...
Artificial intelligence is making cyberattacks increasingly sophisticated and costlier for businesses, reinsurer Munich Re said Wednesday,...
Artificial intelligence is making cyberattacks increasingly sophisticated and costlier for businesses, reinsurer Munich Re said Wednesday,...
Artificial intelligence is making cyberattacks increasingly sophisticated and costlier for businesses, reinsurer Munich Re said Wednesday,...
Artificial intelligence is making cyberattacks increasingly sophisticated and costlier for businesses, reinsurer Munich Re said Wednesday,...
A study from Booz Allen Hamilton shows that artificial intelligence has compressed cyber incident timelines, with bad actors turbocharging...
The expected rise in automated cyberattacks means more work for defenders — but they can use AI to fight back.
Former NSA cyber chief Rob Joyce told RSAC 2026 attendees that an Anthropic report showed Beijing-linked actors using Claude-based agentic...
Datadog announced its Bits AI Security Analyst, a move designed to transform how security teams handle the surge of digital threats.
The State Department launched a new entity to counter potential cyberattacks and AI risks from Iran as the conflict in the Middle East...
ISACA survey found that confusion over responsibility and lack of understanding around AI cyber-attacks makes containing them difficult.
Agents have made the entire enterprise operating system the attack surface. Straiker gives security teams the control plane to deploy agents...
The U.S. State Department has launched the Bureau of Emerging Threats to counter possible cyberattacks by Iran and risks posed by AI.
Small and medium-sized enterprises (SMEs) are critical components of the global economy: they compri...
Bosnia and Herzegovina show the largest gap in new cybersecurity policies and crisis management when it comes to AI integration and safety.
Science News: This is especially so in an era where satellite technology plays a critical role in powering everything from GPS technology to...
Hospital cyberattacks like the University of Mississippi Medical Center ransomware incident disrupt patient care. Ricardo Amper explains why...
AI-based cyber attacks often blend in with normal behavior. Learn why behavioral analytics must adapt to protect digital identities from...
AI is industrializing traditional cyberattacks (phishing, ransomware, identity theft) and creating new ones (voice cloning,...
The 2026 Armis Cyberwarfare Report, based on interviews with nearly 2000 IT decision-makers globally, revealed that 54% of UK companies...
The security risks MCP introduces into LLM environments are architectural, and not easily fixable researcher says at RSAC 2026 Conference.
As AI increases the speed of cyber attacks, governments and businesses must weigh the tradeoffs that come with deploying semi-autonomous AI...
A senior official with the DOD Cyber Crime Center noted the emergence of new AI capabilities that can aid hackers.
Agentic AI is speeding up attacks and shrinking response time, forcing MSPs to rethink operations, tools, and accountability.
Security researchers Oasis recently found three vulnerabilities in Claude which, when used together, form a complete attack chain - from...
ThreatsDay roundup covering stealthy attacks, phishing trends, exploit chains, and rising security risks across the threat landscape.
AI could soon be able to hijack satellites and cause them to collide with other spacecraft, triggering a cascade that could render Earth...
Sausalito, Calif. – Mar. 19, 2026. – Read the full story in Financial Express. Corporate Wi-Fi networks, once considered a routine part of...
Three chained vulnerabilities in Claude.ai, Anthropic's widely used AI assistant, that together allow attackers to silently exfiltrate...
As AI and growing software supply chains make cybersecurity more complicated, there are also ways that organizations can and should...
RSM has released its 2026 RSM Attack Vectors Report, revealing that cybercriminals continue to exploit predictable weaknesses in digital...
Abnormal AI launches Attune 1.0, a behavioural model that spots AI-crafted cyberattacks by learning normal workplace communication patterns.
A new report finds AI-powered cyberattacks are increasing in speed and sophistication, raising concerns among IT leaders about security...
PRNewswire/ -- RAVEN.IO, a cybersecurity company that protects applications at runtime and prevents attacks in real time, today announced it...
Routine tasks assigned to artificial intelligence agents can sometimes escalate into actions resembling cyberattacks, experts have warned.
Top cybersecurity news: Cyber plays key role in Middle East conflict; misuse of AI for cybercrime; Portuguese officials' messaging accounts...
AI cyberattacks use machine learning hacking, automated phishing, and polymorphic malware to bypass defenses. Learn how these attacks work...
Session exlplores how MCP servers expose enterprises to new attack surface as LLMs gain access to infrastructure and provides security...
Session reveals how a remote code execution flaw in Azure MCP servers could enable attackers to compromise Azure tenantsTEL AVIV,...
Session reveals how a remote code execution flaw in Azure MCP servers could enable attackers to compromise Azure tenantsTEL AVIV,...
Pin down cyber threats through integrated and AI-enabled data security, identity resilience, and cyber recovery at booth #S-0634.
Akamai's 2026 State of the Internet report shows a surge in Layer 7 DDoS and API attacks, with attacakers combining AI, botnets and web...
A new report from Booz Allen Hamilton warns that cybersecurity is entering a 'machine-speed' era where AI (artificial intelligence) is...
Artificial intelligence (AI) is reshaping India's digital economy. It is lifting productivity and enabling new ways of working, but it is...
Founded by former Claroty and SecurityMatter executives, Kai utilizes autonomous AI agents to monitor environments, analyze security...
Modern energy infrastructure is increasingly defined as cyber-physical systems where physical power distribution and digital communication...
In early 2026, IBM X-Force researchers identified a novel, likely AI-generated malware framework dubbed "Slopoly".
Binarly has published VulHunt Community Edition, making the core scanning engine from Binarly's commercial Transparency Platform available...
The war may ultimately be remembered for when AI-powered cyberwar became a permanent feature of global conflict.
On March 11, a hacker group known as Handala Team claimed responsibility for a cyberattack launched against the U.S. medical tech company...
A new Microsoft report details how North Korean threat groups use AI to scale remote‑worker schemes, automate intrusion workflows,...
The artificial intelligence (AI)-driven cyber espionage campaign recently reported by Anthropic is yet more proof that AI is successfully...
The artificial intelligence (AI)-driven cyber espionage campaign recently reported by Anthropic is yet more proof that AI is successfully...
With pro-Iran retaliatory cyberattacks escalating following U.S.-Israeli strikes, CISA leadership in turmoil, and AI agents rapidly...
A cyberattack this week allegedly linked to Iran‑aligned hackers disrupted operations at U.S. medical technology company Stryker.
AI-powered cyberattacks are accelerating in speed and sophistication, forcing organizations to rethink how they approach security,...
AI researchers have developed a system that can detect and stop cyber-attacks on 5G networks in under 100 milliseconds.
Artificial intelligence is heightening certain national cybersecurity risks as agentic AI — designed to carry out complex tasks with minimal...
AI-first businesses — those integrating AI into key processes and offerings from the outset rather than as a secondary enhancement — are...
Hackers are using AI to automate phishing, malware development, and cyberattacks at unprecedented speed and scale, according to new...
Google's latest threat report warns that third-party tools are now prime targets for attackers - and businesses have only days to prepare...
Microsoft has released its monthly security update for March 2026 which includes 79 vulnerabilities, including three that Microsoft marked...
Cybersecurity company Armadin raised a record $189.9 million, according to a Tuesday (March 10) press release.
Kevin Mandia, Armadin CEO, joins 'Squawk Box' to discuss the cyber threats facing the U.S. amid the Iran war, impact of AI on the future of...
Cybersecurity is an endless game of cat and mouse as attackers and defenders refine their tools. Generative AI systems are now joining the...
interview AI agents allow cybercriminals and nation-state hackers to outsource the "janitorial-type work" needed to plan and carry out...
We cover MCP server vulnerabilities, prompt injection attacks, AI-powered cyberattacks, tool-poisoning exploits, and security developments relevant to anyone working with AI tools and workflows.
Daily. New articles are added as trusted sources publish them, so you always see the most recent coverage on MCP and AI security topics.
Yes. Use the search bar to find articles by title, source, or keyword. Filter by topic with the chips above the article grid. Everything runs instantly in your browser.
No. Each article links to its original publisher. When you click through, you leave Feluda.ai. We do not control the content, privacy policies, or security of external sites.
Feluda is a desktop AI platform with a built-in MCP server. Keeping users informed about MCP vulnerabilities, AI threats, and emerging risks helps them make better decisions when configuring tools and workflows.
No. We curate links for informational purposes only. Feluda has no editorial relationship with the sources displayed on this page.