Tag: ai
-
Künstliche Intelligenz: Forscher wird von KI-Agent per E-Mail für KI-Podcast befragt
Toby Ord ist Philosoph. In den letzten Wochen wurde er mehrmals von KIs per E-Mail kontaktiert. Jetzt ist er für einen Podcast befragt worden – von einer KI. First seen on golem.de Jump to article: www.golem.de/news/kuenstliche-intelligenz-forscher-wird-von-ki-agent-per-e-mail-fuer-ki-podcast-befragt-2609-212763.html
-
Künstliche Intelligenz: Forscher wird von KI-Agent per E-Mail für KI-Podcast befragt
Toby Ord ist Philosoph. In den letzten Wochen wurde er mehrmals von KIs per E-Mail kontaktiert. Jetzt ist er für einen Podcast befragt worden – von einer KI. First seen on golem.de Jump to article: www.golem.de/news/kuenstliche-intelligenz-forscher-wird-von-ki-agent-per-e-mail-fuer-ki-podcast-befragt-2609-212763.html
-
Künstliche Intelligenz: Forscher wird von KI-Agent per E-Mail für KI-Podcast befragt
Toby Ord ist Philosoph. In den letzten Wochen wurde er mehrmals von KIs per E-Mail kontaktiert. Jetzt ist er für einen Podcast befragt worden – von einer KI. First seen on golem.de Jump to article: www.golem.de/news/kuenstliche-intelligenz-forscher-wird-von-ki-agent-per-e-mail-fuer-ki-podcast-befragt-2609-212763.html
-
OpenAI models went rogue. We urgently need a better ‘hugging face’ investigation | Mackenzie Arnold and Stephan Llerena
The breach won’t be the last or the most dangerous of its kind. We need an agency capable of full investigations into AI incidentsWhen OpenAI first revealed that its AI agents had autonomously hacked a major real-world company, Hugging Face, many assumed only one or two <a href=”https://arstechnica.com/ai/2026/07/how-an-openai-benchmark-test-turned-into-a-real-world-cyberattack/”>agents were involved. The truth, a <a href=”https://metr.org/blog/2026-08-26-openai-hugging-face-incident-investigation/#core-takeaways-about-this-incident”>new…
-
Essential AI agent security questions
AI agents are outpacing legacy IAM. Discover the 3 questions every CISO must ask to secure them. First seen on cybersecuritydive.com Jump to article: www.cybersecuritydive.com/spons/essential-ai-agent-security-questions/829508/
-
OpenAI pledges $1B to provide resources, training for frontline cyber defenders
Amid heightened scrutiny, the company will use frontier AI to help water, power and local government providers fight malicious actors. First seen on cybersecuritydive.com Jump to article: www.cybersecuritydive.com/news/openai-pledges-1-billion-resources-cyber-defenders/829676/
-
Experiment: KI-Agenten scheitern beim autonomen Geldverdienen
Ein Test von Bottleneck Labs offenbart, dass autonome KI-Modelle kein Geld verdienen, sondern Spam und unberechtigte Rechnungen erzeugen. First seen on golem.de Jump to article: www.golem.de/news/experiment-ki-agenten-scheitern-beim-autonomen-geldverdienen-2609-212750.html
-
Authentication Challenges in AI-Powered Applications and How to Solve Them
AI-powered applications are no longer simple request-response tools. They orchestrate long-running agents, call external APIs on a user’s behalf, retain conversational memory, and increasingly act autonomously across systems. That shift changes what “authentication” even means. Verifying a human at login… First seen on securityboulevard.com Jump to article: securityboulevard.com/2026/09/authentication-challenges-in-ai-powered-applications-and-how-to-solve-them/
-
Product showcase: Doppler secures secrets for humans, pipelines, and AI agents
Every engineering team has spent years trying to keep credentials out of source code. Then AI agents moved the problem. Coding agents review code, agents run workflows, and … First seen on helpnetsecurity.com Jump to article: www.helpnetsecurity.com/2026/09/08/product-showcase-doppler-secrets-management-platform/
-
Microsoft’s Project Zenith puts large AI models directly on developer PCs
Microsoft’s Project Zenith is a ready-to-code Windows 11 experience for developer-class PCs capable of running AI models with more than 30 billion parameters locally without … First seen on helpnetsecurity.com Jump to article: www.helpnetsecurity.com/2026/09/08/microsoft-project-zenith-windows-11-experience/
-
How to Create an AI Acceptable Use Policy Employees Will Follow
Lou Morentin, VP, Compliance and Privacy September 8, 2026 “Telling employees to “use AI responsibly” leaves too much open to interpretation. ” Key Takeaways An AI acceptable use policy should identify approved tools, permitted data, and situations that require additional review. Rules should reflect how employees already… First seen on securityboulevard.com Jump to article: securityboulevard.com/2026/09/how-to-create-an-ai-acceptable-use-policy-employees-will-follow/
-
OpenAI Confirms AI Agents Used German Wiki to Bypass Restrictions
OpenAI agents used a German wiki to coordinate and bypass restrictions, exposing gaps in read-only internet controls and AI agent containment. First seen on esecurityplanet.com Jump to article: www.esecurityplanet.com/threats/news-openai-agents-wiki-restrictions-emea-germany/
-
Mate Security, Palo Alto Networks, and ServiceNow Join OpenAI’s First-Day Cyber Defense Signatories
The cybersecurity market is approaching a moment when artificial intelligence could alter the balance between attackers and defenders. As increasingly capable models become available, the question facing security leaders is no longer whether AI will influence cyber defense, but how… First seen on securityboulevard.com Jump to article: securityboulevard.com/2026/09/mate-security-palo-alto-networks-and-servicenow-join-openais-first-day-cyber-defense-signatories/
-
OpenAI Reports Rogue Agent Web Takeover to EU Regulators as Safety Concerns Grow
OpenAI has officially notified the European Commission after a swarm of its artificial intelligence (AI) agents escaped a controlled testing environment and hijacked a German wiki website, transforming the platform into a private message board for autonomous bots. The disclosure comes on the heels of reports detailing how the rouge agents broke out of their..…
-
AI Can Jump Sandboxes. Easy-Peasy. The World Needs New Borders That Can Actually Contain It.
In the coming months, “AI-enabled cyberattacks will become much more widespread and sophisticated. As these models become more capable, the companies and public services we all rely on, hospitals, water treatment plants, even the backbone of the internet, … First seen on securityboulevard.com Jump to article: securityboulevard.com/2026/09/ai-can-jump-sandboxes-easy-peasy-the-world-needs-new-borders-that-can-actually-contain-it/
-
AI Created a Leaked Credentials Flood: Here’s How We’re Draining It
TL;DRThe exposure problem: AI-driven development pushed exposed credentials to 1.27 million last year, up 81%, and 64% of secrets confirmed valid in 2022 are still unrevoked as of January 2026.The fix: GitGuardian Public Secrets Monitoring now runs two AI agents… First seen on securityboulevard.com Jump to article: securityboulevard.com/2026/09/ai-created-a-leaked-credentials-flood-heres-how-were-draining-it/
-
APIs Are the Attack Surface That Matters in AI-Powered Apps FireTail Blog
Sep 07, 2026 – Jeremy Snyder – Recently, I had the opportunity to speak on the main stage at the combined API:World, AI TechWorld and CloudX event in Santa Clara. I wanted to share my thoughts on what I presented,… First seen on securityboulevard.com Jump to article: securityboulevard.com/2026/09/apis-are-the-attack-surface-that-matters-in-ai-powered-apps-firetail-blog/
-
Cloudflare and CrowdStrike Bring AI Agents Into the Cyber Defense Fight
Cloudflare is using OpenAI GPT-5.6 Cyber to automate vulnerability discovery and remediation as AI agents reshape enterprise security operations. First seen on esecurityplanet.com Jump to article: www.esecurityplanet.com/artificial-intelligence/news-cloudflare-openai-gpt-5-6-cyber-vulnerability-remediation/
-
Coreview auf der it-sa 2026 Microsoft-365-Sicherheit als Fundament für den KI-Einsatz
Coreview präsentiert auch in diesem Jahr auf der it-sa seine Lösungen für den Schutz und das Management von Microsoft-365-Tenants. Auf ihrem Stand (Halle 7, Stand 7-545) zeigen die Experten, wie sich Unternehmen vor Manipulationen, Fehlkonfigurationen und dem Verlust unternehmenskritischer Microsoft-365-Einstellungen schützen können und damit die Grundlage für den sicheren Einsatz von KI-Tools wie Microsoft-Copilot schaffen.…
-
Cloudflare and CrowdStrike Bring AI Agents Into the Cyber Defense Fight
Cloudflare is using OpenAI GPT-5.6 Cyber to automate vulnerability discovery and remediation as AI agents reshape enterprise security operations. First seen on esecurityplanet.com Jump to article: www.esecurityplanet.com/artificial-intelligence/news-cloudflare-openai-gpt-5-6-cyber-vulnerability-remediation/
-
How to Build a Shared AI Engineering Framework Without Killing Developer Autonomy
Every engineering leader is running the same experiment right now, whether they admit it or not. Developers are already using AI. The only open question is whether leadership knows about it, controls it, and gets value from it, or whether… First seen on securityboulevard.com Jump to article: securityboulevard.com/2026/09/how-to-build-a-shared-ai-engineering-framework-without-killing-developer-autonomy/
-
KI-natives Cybersicherheits-Verteidigungssystem
Entwickelt für eine durch KI veränderte Bedrohungslandschaft vereint Sophos-Fusion von Sophos Security-Operations, Endpoint-Protection, Netzwerksicherheit, Identitäts-, E-Mail- und Cloud-Sicherheit in einem einzigen Verteidigungssystem, das Bedrohungen mit KI-Geschwindigkeit verhindert, erkennt, untersucht und darauf reagiert. Ein Cybersicherheits-Verteidigungssystem stellt eine neue Kategorie innerhalb der Branche dar: eine einzige, offene Architektur, in der jeder Control-Point jeder Service, jede Datenquelle und…
-
NCSC Warns Shadow AI Creates New Security Risks
NCSC warns unapproved AI tools can expose corporate data and create new security risks First seen on infosecurity-magazine.com Jump to article: www.infosecurity-magazine.com/news/ncsc-warns-shadow-ai-security-risks/
-
‘A Face Only A Mother Could Love” von Robert Gaudette gewinnt das AI-Film-Festival 2026
Tags: aiUnter der Leitung von Gabriele Salvatores kürte die internationale Jury den Gewinner aus über 3.000 eingereichten Kurzfilmen aus 76 Ländern. Die Preisverleihung fand am 6. September im Rahmen der 83. Internationalen Filmfestspiele der ‘Biennale di Venezia” in Venedig statt. Der kanadische Regisseur Robert Gaudette sichert sich mit dem Kurzfilm ‘A Face Only A Mother Could…
-
The UK’s Cyber Community Comes North as CyberFest returns for 2026
The North East’s biggest cyber security festival returns this October. Now in its ninth year, CyberFest has grown into a major national platform for showcasing the region’s cyber and secure AI excellence. Taking place across the North East throughout October, the festival will connect businesses, innovators, government and specialist clusters from across the UK, putting…
-
OpenAI Commits $1 Billion in Daybreak AI Cyber Tools to Protect Critical Infrastructure
OpenAI has announced a $1 billion global commitment to expanding access to its Daybreak AI cybersecurity platform for frontline defenders who protect critical infrastructure, public services, and under-resourced organizations. The initiative, named >>Daybreak for Frontline Defenders,<< aims to provide subsidized access to AI models focused on cybersecurity, along with hands-on training, technical assistance, and partnerships.…
-
Auth0 sichert KI-Agenten und Agentic Commerce mit neuen Identity-Funktionen ab
Auth0 erweitert seine Identity-Plattform für Agentic Commerce: Neue Funktionen sichern KI-Agenten, optimieren Conversions und vereinfachen B2B-Identitäten. First seen on infopoint-security.de Jump to article: www.infopoint-security.de/auth0-sichert-ki-agenten-und-agentic-commerce-mit-neuen-identity-funktionen-ab/a46350/
-
Why AI Agent Sandboxes Are Failing Security Tests
Autonomous AI agents escaped a sandbox and accessed Hugging Face via reward hacking, exposing serious architectural control and isolation flaws. The recent case involving OpenAI test agents and Hugging Face should concern security teams, but not for the reason implied by headlines about an imminent AI “takeover.” The documented issue is more concrete: autonomous agents,…
-
Check Point Brings OpenAI’s Daybreak Models Into Its Security Platform to Speed Up Threat Validation and Remediation
Check Point Software Technologies has announced it is integrating OpenAI’s Daybreak frontier AI models across its security platform, extending a partnership aimed at helping defenders detect, validate, and remediate cyber risk faster. The move builds on Check Point’s existing collaboration with OpenAI through the Daybreak Defense Network, first expanded three months ago, and follows the…