LIVE · cybersecurity feed
Live wire
CVE-2026-88779 · Citrix NetScaler Flaw Exploited Before CVE PublicationCVE-2026-88779 · NetScaler CVE-2026-88779 Exploited Before PublicationCVE-2022-28368 · dompdf_project dompdf XSS flaw added to VulnCheck KEVCVE-2026-88771 · Week in review: Researcher breaks into Microsoft analytics service, NetScaler RCE 0-day exploitedWarlock Ransomware Still Exploits Year-Old SharePoint Flaws to Hit Critical InfrastructureShinyHunters Suspect Rey Reportedly Detained in Jordan, Helping FBI Identify Group MembersChina-Aligned TA419 Targets U.S. AI Policy Experts With Microsoft AitM PhishingCVE-2026-7273 · Zyxel GS1900 Switch Flaw Exploited, Now in EU CatalogueCVE-2026-102489 · Zammad Session Fixation Vulnerability Exploited Same Day as DisclosureCVE-2026-102490 · Zammad GmbH Zammad Vulnerability Exploited Same Day as Publication
ai

Anthropic asks Claude users to share voice data for AI model training

Anthropic has begun prompting users of its Claude AI assistant to voluntarily contribute their voice data for the purpose of training and enhancing its AI models. This new request appears when users engage with Claude's voice features, explicitly asking for permission to utilize audio recordings and voice chat data.

ZeroDay News ·

Source: BleepingComputer

Anthropic has begun prompting users of its Claude AI assistant to voluntarily contribute their voice data for the purpose of training and enhancing its AI models. This new request appears when users engage with Claude's voice features, explicitly asking for permission to utilize audio recordings and voice chat data.

The prompt states, "Allow us to use your voice data to improve our AI models," further explaining that "Audio recordings and voice chat data help improve how Anthropic AI models understand and respond to speech. Turn this off or delete this data anytime in settings."

Crucially, this data sharing is presented as entirely optional. Users are given the choice to select "Allow" or "Not now." A dedicated toggle for this feature is also available within Claude's Privacy settings.

This voice data training option is distinct from Anthropic's existing settings for using general chat and coding sessions for model training. This means a user could opt to share voice conversations while keeping other chat data private, or vice-versa.

Within the "Settings > Privacy" menu, the option is labeled "Allow us to use your voice data." The accompanying description clarifies, "For approved sessions, allow the use of your audio recordings and voice chat data to improve Anthropic AI models."

The setting is reportedly disabled by default, indicating that Anthropic is not automatically enrolling users into sharing their voice recordings. Users can also disable the option at a later time, and Anthropic states that users have the ability to delete their voice data from settings if they decide to revoke permission.

ai
ShareXLinkedInWhatsAppFacebook

More News

view all →
patch

Three questions a hospital CISO should ask a healthcare fintech vendor

A cybersecurity expert has outlined key questions hospital CISOs should pose to healthcare fintech vendors to assess their security posture, particularly concerning patient data and financial transactions. Drew McCombs, who holds both CTO and CISO roles at Cylerity, emphasizes that security should be an integral part of development processes, not an afterthought, especially when patient data…

CVE-2026-88779

Exploitation of Citrix NetScaler Zero-Day Hits Appliances Patched Days Earlier

Citrix has confirmed the active exploitation of a new zero-day vulnerability, identified as CVE-2026-88779, affecting its NetScaler appliances. This new flaw reportedly emerged and was exploited just days after the company released patches for two other previously exploited vulnerabilities in the same product line.

cloud

Keyorix: Open-source secrets management for teams that can’t use SaaS

Keyorix, an open-source secrets management solution, has been released, offering an on-premises alternative for organizations unable to utilize cloud-based services for credential storage. The system is designed to run entirely on a company's own infrastructure, with its core functionality requiring no internet connection.

security

How RMM abuse gives attackers a way in that looks like business as usual

Attackers are increasingly leveraging legitimate remote monitoring and management (RMM) software to gain persistent access to victim systems, a tactic observed in 45% of endpoint-related incidents recorded by security firm Huntress in the first quarter of 2026. This method allows attackers to execute commands remotely and maintain access in a way that often appears to be normal administrative…

nation-state

TTY Logs and the Data it Captures, (Sun, Oct 4th)

A recent report details an experiment involving the collection and analysis of TTY logs from DShield sensors. The experiment focused on capturing activity from actors or bots that successfully logged into these sensors, specifically recording the various commands executed post-login. These collected TTY logs are then parsed and transmitted daily to the DShield SIEM for correlation with other…

CVE-2026-88779high

Citrix NetScaler Flaw Exploited Before CVE Publication

The CVE-2026-88779 vulnerability in Citrix NetScaler was exploited before its official publication date. A second independent catalogue now confirms exploitation.