Tag: api

Whisper Leak uses a side channel attack to eavesdrop on encrypted AI conversations

Nov 10, 2025 2:19 PM

Tags: ai, api, attack, ciso, cloud, cyberattack, data, finance, healthcare, leak, LLM, microsoft, mitigation, network, openai, service, side-channel, vpn

Inside Microsoft’s proof-of-concept: Researchers at Microsoft simulated a real-world scenario in which the adversary could observe encrypted traffic but not decrypt it. They chose “legality of money laundering” as the target topic for the proof-of-concept.For positive samples, the team used a”¯language model”¯to generate 100 semantically similar variants of questions about this topic. For negative”¯noise”¯samples,”¯it randomly…
Whisper Leak uses a side channel attack to eavesdrop on encrypted AI conversations

Nov 10, 2025 2:19 PM

Tags: ai, api, attack, ciso, cloud, cyberattack, data, finance, healthcare, leak, LLM, microsoft, mitigation, network, openai, service, side-channel, vpn

Inside Microsoft’s proof-of-concept: Researchers at Microsoft simulated a real-world scenario in which the adversary could observe encrypted traffic but not decrypt it. They chose “legality of money laundering” as the target topic for the proof-of-concept.For positive samples, the team used a”¯language model”¯to generate 100 semantically similar variants of questions about this topic. For negative”¯noise”¯samples,”¯it randomly…
Whisper Leak uses a side channel attack to eavesdrop on encrypted AI conversations

Nov 10, 2025 2:19 PM

Tags: ai, api, attack, ciso, cloud, cyberattack, data, finance, healthcare, leak, LLM, microsoft, mitigation, network, openai, service, side-channel, vpn

Inside Microsoft’s proof-of-concept: Researchers at Microsoft simulated a real-world scenario in which the adversary could observe encrypted traffic but not decrypt it. They chose “legality of money laundering” as the target topic for the proof-of-concept.For positive samples, the team used a”¯language model”¯to generate 100 semantically similar variants of questions about this topic. For negative”¯noise”¯samples,”¯it randomly…
Whisper Leak uses a side channel attack to eavesdrop on encrypted AI conversations

Nov 10, 2025 2:19 PM

Tags: ai, api, attack, ciso, cloud, cyberattack, data, finance, healthcare, leak, LLM, microsoft, mitigation, network, openai, service, side-channel, vpn

Inside Microsoft’s proof-of-concept: Researchers at Microsoft simulated a real-world scenario in which the adversary could observe encrypted traffic but not decrypt it. They chose “legality of money laundering” as the target topic for the proof-of-concept.For positive samples, the team used a”¯language model”¯to generate 100 semantically similar variants of questions about this topic. For negative”¯noise”¯samples,”¯it randomly…
SECURITY AFFAIRS MALWARE NEWSLETTER ROUND 70

Nov 9, 2025 12:20 PM

Tags: api, backdoor, control, defense, international, malware, military, openai

Security Affairs Malware newsletter includes a collection of the best articles and research on malware in the international landscape Malware Newsletter SesameOp: Novel backdoor uses OpenAI Assistants API for command and control Weaponized Military Documents Deliver Advanced SSH-Tor Backdoor to Defense Sector Gootloader Returns: What Goodies Did They Bring? Ransomvibing appears in VS Code extensions…
NHIs: A Budget-Friendly Solution for Modern Cybersecurity?

Nov 9, 2025 5:19 AM

Tags: api, cloud, cybersecurity, iot

How Are Non-Human Identities Reshaping Cybersecurity Solutions? How often have you considered the role of machine identities? With more enterprises migrate their operations to the cloud, the management of Non-Human Identities (NHIs) has become paramount. These machine identities, which include everything from APIs to IoT devices, are proving to be pivotal in maintaining robust cybersecurity……
NHIs: A Budget-Friendly Solution for Modern Cybersecurity?

Nov 9, 2025 5:19 AM

Tags: api, cloud, cybersecurity, iot

How Are Non-Human Identities Reshaping Cybersecurity Solutions? How often have you considered the role of machine identities? With more enterprises migrate their operations to the cloud, the management of Non-Human Identities (NHIs) has become paramount. These machine identities, which include everything from APIs to IoT devices, are proving to be pivotal in maintaining robust cybersecurity……
Cybersecurity Snapshot: AI Will Take Center Stage in Cyber in 2026, Google Says, as MITRE Revamps ATTCK Framework

Nov 7, 2025 8:20 PM

Tags: access, ai, android, antivirus, api, apple, attack, authentication, business, cio, cisa, ciso, cloud, communications, container, control, credentials, cryptography, cve, cyber, cybercrime, cybersecurity, data, defense, detection, docker, endpoint, exploit, firewall, flaw, framework, google, governance, group, guide, hacker, identity, infrastructure, injection, Internet, kubernetes, leak, least-privilege, linux, malicious, malware, mfa, microsoft, mitigation, mitre, mobile, network, offense, oracle, programming, resilience, risk, risk-management, service, skills, soc, social-engineering, software, sophos, spam, strategy, supply-chain, switch, tactics, technology, threat, tool, unauthorized, update, vulnerability, windows, zero-day

Learn why Google expects AI to transform cyber defense and offense next year, and explore MITRE’s major update to the ATT&CK knowledge base. We also cover a new McKinsey playbook for agentic AI security, along with the latest on Microsoft Exchange protection and the CIS Benchmarks. Key takeaways Google is forecasting that AI will kick…
Cybersecurity Snapshot: AI Will Take Center Stage in Cyber in 2026, Google Says, as MITRE Revamps ATTCK Framework

Nov 7, 2025 8:20 PM

Tags: access, ai, android, antivirus, api, apple, attack, authentication, business, cio, cisa, ciso, cloud, communications, container, control, credentials, cryptography, cve, cyber, cybercrime, cybersecurity, data, defense, detection, docker, endpoint, exploit, firewall, flaw, framework, google, governance, group, guide, hacker, identity, infrastructure, injection, Internet, kubernetes, leak, least-privilege, linux, malicious, malware, mfa, microsoft, mitigation, mitre, mobile, network, offense, oracle, programming, resilience, risk, risk-management, service, skills, soc, social-engineering, software, sophos, spam, strategy, supply-chain, switch, tactics, technology, threat, tool, unauthorized, update, vulnerability, windows, zero-day

Learn why Google expects AI to transform cyber defense and offense next year, and explore MITRE’s major update to the ATT&CK knowledge base. We also cover a new McKinsey playbook for agentic AI security, along with the latest on Microsoft Exchange protection and the CIS Benchmarks. Key takeaways Google is forecasting that AI will kick…
KI-Malware ist keine Theorie mehr

Nov 7, 2025 2:20 PM

Tags: access, ai, antivirus, api, control, cyberattack, cybercrime, cybersecurity, data, exploit, github, google, group, hacker, intelligence, LLM, malware, ransomware, RedTeam, service, skills, social-engineering, software, threat, tool, vulnerability

KI boomt auch unter Cyberkriminellen. Die ersten operativen Ergebnisse dieses Trends beleuchten Google-Sicherheitsforscher in einem aktuellen Report.Was lange befürchtet und vermutet wurde, will die Google Threat Intelligence Group (GTIG) nun im Rahmen einer aktuellen Untersuchung belegen: Cyberkriminelle nutzen KI im Rahmen ihrer Malware-Angriffskampagnen. Aber längst nicht mehr nur für Vibe-Coding-Zwecke oder zur technischen Unterstützung. Wie…
Flaw in React Native CLI opens dev servers to attacks

Nov 7, 2025 2:20 PM

Tags: access, api, attack, flaw, malicious, network, tool, vulnerability

What developers must do now?: Developers using @react-native-community/cli (or the bundled cli-server-api) in their React Native projects should check for the vulnerable package version on the npm list. The vulnerability is fixed in version 20.0.0 of cli-server-api, so immediate updating is recommended.The stakes include an attacker remotely executing commands on the victim’s development machine, potentially…
KI-Malware ist keine Theorie mehr

Nov 7, 2025 2:20 PM

Tags: access, ai, antivirus, api, control, cyberattack, cybercrime, cybersecurity, data, exploit, github, google, group, hacker, intelligence, LLM, malware, ransomware, RedTeam, service, skills, social-engineering, software, threat, tool, vulnerability

KI boomt auch unter Cyberkriminellen. Die ersten operativen Ergebnisse dieses Trends beleuchten Google-Sicherheitsforscher in einem aktuellen Report.Was lange befürchtet und vermutet wurde, will die Google Threat Intelligence Group (GTIG) nun im Rahmen einer aktuellen Untersuchung belegen: Cyberkriminelle nutzen KI im Rahmen ihrer Malware-Angriffskampagnen. Aber längst nicht mehr nur für Vibe-Coding-Zwecke oder zur technischen Unterstützung. Wie…
Flaw in React Native CLI opens dev servers to attacks

Nov 7, 2025 2:20 PM

Tags: access, api, attack, flaw, malicious, network, tool, vulnerability

What developers must do now?: Developers using @react-native-community/cli (or the bundled cli-server-api) in their React Native projects should check for the vulnerable package version on the npm list. The vulnerability is fixed in version 20.0.0 of cli-server-api, so immediate updating is recommended.The stakes include an attacker remotely executing commands on the victim’s development machine, potentially…
9 top bug bounty programs launched in 2025

Nov 7, 2025 2:20 PM

Tags: ai, api, apple, attack, best-practice, browser, bug-bounty, chatgpt, chrome, cloud, computer, corporate, cyber, cybersecurity, data, defense, exploit, finance, flaw, framework, google, government, hacker, hacking, healthcare, identity, infrastructure, injection, intelligence, iphone, macOS, microsoft, mitigation, mobile, nis-2, nvidia, openai, phishing, risk, service, spyware, strategy, technology, theft, threat, unauthorized, update, vulnerability, zero-day

Bug bounty programs focus in 2025: For ethical hackers, best practice for bug bounty hunting in 2025 involves thorough reconnaissance of a target organization’s technology stack, rather than just running automated tools.Leading bug bounty platforms such as Bugcrowd, HackerOne, Synack, YesWeHack, and Intigriti offer rewards for identifying and reporting security vulnerabilities. Platforms connect ethical hackers…
Google researchers detect first operational use of LLMs in active malware campaigns

Nov 7, 2025 2:20 PM

Tags: ai, api, attack, cybercrime, cybersecurity, encryption, exploit, finance, google, group, iran, LLM, malware, marketplace, phishing, RedTeam, service, skills, social-engineering, threat, tool, vulnerability

Using social engineering against LLMs: Additionally, GTIG found that attackers are increasingly using “social engineering-like pretexts” in their prompts to get around LLM safeguards. Notably, they have posed as participants in a “capture-the-flag” (CTF) gamified cybersecurity competition, persuading Gemini to give up information it would otherwise refuse to reveal. In one interaction, for instance, an attacker…
9 top bug bounty programs launched in 2025

Nov 7, 2025 2:20 PM

Tags: ai, api, apple, attack, best-practice, browser, bug-bounty, chatgpt, chrome, cloud, computer, corporate, cyber, cybersecurity, data, defense, exploit, finance, flaw, framework, google, government, hacker, hacking, healthcare, identity, infrastructure, injection, intelligence, iphone, macOS, microsoft, mitigation, mobile, nis-2, nvidia, openai, phishing, risk, service, spyware, strategy, technology, theft, threat, unauthorized, update, vulnerability, zero-day

Bug bounty programs focus in 2025: For ethical hackers, best practice for bug bounty hunting in 2025 involves thorough reconnaissance of a target organization’s technology stack, rather than just running automated tools.Leading bug bounty platforms such as Bugcrowd, HackerOne, Synack, YesWeHack, and Intigriti offer rewards for identifying and reporting security vulnerabilities. Platforms connect ethical hackers…
Google researchers detect first operational use of LLMs in active malware campaigns

Nov 7, 2025 2:20 PM

Tags: ai, api, attack, cybercrime, cybersecurity, encryption, exploit, finance, google, group, iran, LLM, malware, marketplace, phishing, RedTeam, service, skills, social-engineering, threat, tool, vulnerability

Using social engineering against LLMs: Additionally, GTIG found that attackers are increasingly using “social engineering-like pretexts” in their prompts to get around LLM safeguards. Notably, they have posed as participants in a “capture-the-flag” (CTF) gamified cybersecurity competition, persuading Gemini to give up information it would otherwise refuse to reveal. In one interaction, for instance, an attacker…
9 top bug bounty programs launched in 2025

Nov 7, 2025 2:20 PM

Tags: ai, api, apple, attack, best-practice, browser, bug-bounty, chatgpt, chrome, cloud, computer, corporate, cyber, cybersecurity, data, defense, exploit, finance, flaw, framework, google, government, hacker, hacking, healthcare, identity, infrastructure, injection, intelligence, iphone, macOS, microsoft, mitigation, mobile, nis-2, nvidia, openai, phishing, risk, service, spyware, strategy, technology, theft, threat, unauthorized, update, vulnerability, zero-day

Bug bounty programs focus in 2025: For ethical hackers, best practice for bug bounty hunting in 2025 involves thorough reconnaissance of a target organization’s technology stack, rather than just running automated tools.Leading bug bounty platforms such as Bugcrowd, HackerOne, Synack, YesWeHack, and Intigriti offer rewards for identifying and reporting security vulnerabilities. Platforms connect ethical hackers…
9 top bug bounty programs launched in 2025

Nov 7, 2025 2:20 PM

Tags: ai, api, apple, attack, best-practice, browser, bug-bounty, chatgpt, chrome, cloud, computer, corporate, cyber, cybersecurity, data, defense, exploit, finance, flaw, framework, google, government, hacker, hacking, healthcare, identity, infrastructure, injection, intelligence, iphone, macOS, microsoft, mitigation, mobile, nis-2, nvidia, openai, phishing, risk, service, spyware, strategy, technology, theft, threat, unauthorized, update, vulnerability, zero-day

Bug bounty programs focus in 2025: For ethical hackers, best practice for bug bounty hunting in 2025 involves thorough reconnaissance of a target organization’s technology stack, rather than just running automated tools.Leading bug bounty platforms such as Bugcrowd, HackerOne, Synack, YesWeHack, and Intigriti offer rewards for identifying and reporting security vulnerabilities. Platforms connect ethical hackers…
9 top bug bounty programs launched in 2025

Nov 7, 2025 2:20 PM

Tags: ai, api, apple, attack, best-practice, browser, bug-bounty, chatgpt, chrome, cloud, computer, corporate, cyber, cybersecurity, data, defense, exploit, finance, flaw, framework, google, government, hacker, hacking, healthcare, identity, infrastructure, injection, intelligence, iphone, macOS, microsoft, mitigation, mobile, nis-2, nvidia, openai, phishing, risk, service, spyware, strategy, technology, theft, threat, unauthorized, update, vulnerability, zero-day

Bug bounty programs focus in 2025: For ethical hackers, best practice for bug bounty hunting in 2025 involves thorough reconnaissance of a target organization’s technology stack, rather than just running automated tools.Leading bug bounty platforms such as Bugcrowd, HackerOne, Synack, YesWeHack, and Intigriti offer rewards for identifying and reporting security vulnerabilities. Platforms connect ethical hackers…
Defending digital identity from computer-using agents (CUAs)

Nov 7, 2025 2:20 PM

Tags: access, ai, api, attack, authentication, automation, breach, captcha, cisa, computer, control, corporate, credentials, cybersecurity, data, data-breach, defense, detection, email, exploit, fido, google, identity, infrastructure, login, malicious, malware, nfc, passkey, password, phishing, saas, service, skills, social-engineering, tool, unauthorized, update

Principle of least effort: Our brains seek shortcuts to reduce cognitive load, making password reuse seem rational.Security fatigue: Frequent password changes and complex rules frustrate users, pushing them toward reuse.As a result, users often rotate between 410 core passwords. According to an article by Enzoic, the average person reuses the same password across as many as 14…
9 top bug bounty programs launched in 2025

Nov 7, 2025 2:20 PM

Tags: ai, api, apple, attack, best-practice, browser, bug-bounty, chatgpt, chrome, cloud, computer, corporate, cyber, cybersecurity, data, defense, exploit, finance, flaw, framework, google, government, hacker, hacking, healthcare, identity, infrastructure, injection, intelligence, iphone, macOS, microsoft, mitigation, mobile, nis-2, nvidia, openai, phishing, risk, service, spyware, strategy, technology, theft, threat, unauthorized, update, vulnerability, zero-day

Bug bounty programs focus in 2025: For ethical hackers, best practice for bug bounty hunting in 2025 involves thorough reconnaissance of a target organization’s technology stack, rather than just running automated tools.Leading bug bounty platforms such as Bugcrowd, HackerOne, Synack, YesWeHack, and Intigriti offer rewards for identifying and reporting security vulnerabilities. Platforms connect ethical hackers…
Defending digital identity from computer-using agents (CUAs)

Nov 7, 2025 2:20 PM

Tags: access, ai, api, attack, authentication, automation, breach, captcha, cisa, computer, control, corporate, credentials, cybersecurity, data, data-breach, defense, detection, email, exploit, fido, google, identity, infrastructure, login, malicious, malware, nfc, passkey, password, phishing, saas, service, skills, social-engineering, tool, unauthorized, update

Principle of least effort: Our brains seek shortcuts to reduce cognitive load, making password reuse seem rational.Security fatigue: Frequent password changes and complex rules frustrate users, pushing them toward reuse.As a result, users often rotate between 410 core passwords. According to an article by Enzoic, the average person reuses the same password across as many as 14…
Flaw in React Native CLI opens dev servers to attacks

Nov 7, 2025 2:20 PM

Tags: access, api, attack, flaw, malicious, network, tool, vulnerability

What developers must do now?: Developers using @react-native-community/cli (or the bundled cli-server-api) in their React Native projects should check for the vulnerable package version on the npm list. The vulnerability is fixed in version 20.0.0 of cli-server-api, so immediate updating is recommended.The stakes include an attacker remotely executing commands on the victim’s development machine, potentially…
Flaw in React Native CLI opens dev servers to attacks

Nov 7, 2025 2:20 PM

Tags: access, api, attack, flaw, malicious, network, tool, vulnerability

What developers must do now?: Developers using @react-native-community/cli (or the bundled cli-server-api) in their React Native projects should check for the vulnerable package version on the npm list. The vulnerability is fixed in version 20.0.0 of cli-server-api, so immediate updating is recommended.The stakes include an attacker remotely executing commands on the victim’s development machine, potentially…
KI-Malware ist keine Theorie mehr

Nov 7, 2025 2:20 PM

Tags: access, ai, antivirus, api, control, cyberattack, cybercrime, cybersecurity, data, exploit, github, google, group, hacker, intelligence, LLM, malware, ransomware, RedTeam, service, skills, social-engineering, software, threat, tool, vulnerability

KI boomt auch unter Cyberkriminellen. Die ersten operativen Ergebnisse dieses Trends beleuchten Google-Sicherheitsforscher in einem aktuellen Report.Was lange befürchtet und vermutet wurde, will die Google Threat Intelligence Group (GTIG) nun im Rahmen einer aktuellen Untersuchung belegen: Cyberkriminelle nutzen KI im Rahmen ihrer Malware-Angriffskampagnen. Aber längst nicht mehr nur für Vibe-Coding-Zwecke oder zur technischen Unterstützung. Wie…
AI is the New Insider Threat: Rethinking Enterprise Security in the Digital Age

Nov 6, 2025 9:19 PM

Tags: access, ai, api, breach, business, cloud, compliance, container, control, data, detection, email, encryption, finance, framework, gartner, governance, group, ibm, identity, intelligence, jobs, malicious, ml, monitoring, resilience, risk, service, software, strategy, technology, threat, tool, training, unauthorized, update, vulnerability, zero-trust

AI is the New Insider Threat: Rethinking Enterprise Security in the Digital Age madhav Thu, 11/06/2025 – 13:02 Artificial intelligence (AI) is no longer just a passive tool. It’s an active insider interpreting data, executing workflows, automating decisions, accessing sensitive data, and managing critical systems, in enterprise operations that directly affects an enterprise risk posture…
AI is the New Insider Threat: Rethinking Enterprise Security in the Digital Age

Nov 6, 2025 9:19 PM

Tags: access, ai, api, breach, business, cloud, compliance, container, control, data, detection, email, encryption, finance, framework, gartner, governance, group, ibm, identity, intelligence, jobs, malicious, ml, monitoring, resilience, risk, service, software, strategy, technology, threat, tool, training, unauthorized, update, vulnerability, zero-trust

AI is the New Insider Threat: Rethinking Enterprise Security in the Digital Age madhav Thu, 11/06/2025 – 13:02 Artificial intelligence (AI) is no longer just a passive tool. It’s an active insider interpreting data, executing workflows, automating decisions, accessing sensitive data, and managing critical systems, in enterprise operations that directly affects an enterprise risk posture…
Do robots dream of secure networking? Teaching cybersecurity to AI systems

Nov 6, 2025 12:19 PM

Tags: ai, api, cisco, cybersecurity, intelligence, openai, threat

This blog demonstrates a proof of concept using LangChain and OpenAI, integrated with Cisco Umbrella API, to provide AI agents with real-time threat intelligence for evaluating domain dispositions. First seen on blog.talosintelligence.com Jump to article: blog.talosintelligence.com/do-robots-dream-of-secure-networking/
Do robots dream of secure networking? Teaching cybersecurity to AI systems

Nov 6, 2025 12:19 PM

Tags: ai, api, cisco, cybersecurity, intelligence, openai, threat

This blog demonstrates a proof of concept using LangChain and OpenAI, integrated with Cisco Umbrella API, to provide AI agents with real-time threat intelligence for evaluating domain dispositions. First seen on blog.talosintelligence.com Jump to article: blog.talosintelligence.com/do-robots-dream-of-secure-networking/