Tag: Act

  • The Register: Microsoft fixes under-attack privilege-escalation holes in Hyper-V

    Source URL: https://www.theregister.com/2025/01/15/patch_tuesday_january_2025/ Source: The Register Title: Microsoft fixes under-attack privilege-escalation holes in Hyper-V Feedly Summary: Plus: Excel hell, angst for Adobe fans, and life’s too Snort for Cisco Patch Tuesday The first Patch Tuesday of 2025 has seen Microsoft address three under-attack privilege-escalation flaws in its Hyper-V hypervisor, plus plenty more problems that deserve…

  • Slashdot: OpenAI’s AI Reasoning Model ‘Thinks’ In Chinese Sometimes, No One Really Knows Why

    Source URL: https://slashdot.org/story/25/01/14/239246/openais-ai-reasoning-model-thinks-in-chinese-sometimes-no-one-really-knows-why?utm_source=rss1.0mainlinkanon&utm_medium=feed Source: Slashdot Title: OpenAI’s AI Reasoning Model ‘Thinks’ In Chinese Sometimes, No One Really Knows Why Feedly Summary: AI Summary and Description: Yes Summary: The behavior exhibited by OpenAI’s reasoning AI model, o1, which seemingly “thinks” in multiple languages regardless of the input language, has raised questions within the AI community. Experts…

  • Hacker News: Don’t use cosine similarity carelessly

    Source URL: https://p.migdal.pl/blog/2025/01/dont-use-cosine-similarity/ Source: Hacker News Title: Don’t use cosine similarity carelessly Feedly Summary: Comments AI Summary and Description: Yes Summary: The text explores the complexities and limitations of using cosine similarity in AI, particularly in the context of vector embeddings derived from language models. It critiques the blind application of cosine similarity to assess…

  • Krebs on Security: Microsoft: Happy 2025. Here’s 161 Security Updates

    Source URL: https://krebsonsecurity.com/2025/01/microsoft-happy-2025-heres-161-security-updates/ Source: Krebs on Security Title: Microsoft: Happy 2025. Here’s 161 Security Updates Feedly Summary: Microsoft today unleashed updates to plug a whopping 161 security vulnerabilities in Windows and related software, including three “zero-day" weaknesses that are already under active attack. Redmond’s inaugural Patch Tuesday of 2025 bundles more fixes than the company…

  • Alerts: CISA Adds Four Known Exploited Vulnerabilities to Catalog

    Source URL: https://www.cisa.gov/news-events/alerts/2025/01/14/cisa-adds-four-known-exploited-vulnerabilities-catalog Source: Alerts Title: CISA Adds Four Known Exploited Vulnerabilities to Catalog Feedly Summary: CISA has added four vulnerabilities to its Known Exploited Vulnerabilities Catalog, based on evidence of active exploitation. CVE-2024-55591 Fortinet FortiOS Authorization Bypass Vulnerability CVE-2025-21333 Microsoft Windows Hyper-V NT Kernel Integration VSP Heap-based Buffer Overflow Vulnerability CVE-2025-21334 Microsoft Windows Hyper-V NT…

  • Slashdot: Texas Sues Allstate For Collecting Driver Data To Raise Premiums

    Source URL: https://tech.slashdot.org/story/25/01/14/2042251/texas-sues-allstate-for-collecting-driver-data-to-raise-premiums?utm_source=rss1.0mainlinkanon&utm_medium=feed Source: Slashdot Title: Texas Sues Allstate For Collecting Driver Data To Raise Premiums Feedly Summary: AI Summary and Description: Yes **Summary:** Texas has initiated a significant lawsuit against Allstate and its subsidiary Arity for allegedly violating the state’s privacy laws by secretly collecting location data from millions of drivers, which was used…

  • The Register: FBI wipes Chinese PlugX malware from thousands of Windows PCs in America

    Source URL: https://www.theregister.com/2025/01/14/fbi_french_cops_boot_chinas/ Source: The Register Title: FBI wipes Chinese PlugX malware from thousands of Windows PCs in America Feedly Summary: Hey, Xi: Zài jiàn! The FBI, working with French cops, obtained nine warrants to remotely wipe PlugX malware from thousands of Windows-based computers that had been infected by Chinese government-backed criminals, according to newly…

  • Hacker News: Show HN: Value likelihoods for OpenAI structured output

    Source URL: https://arena-ai.github.io/structured-logprobs/ Source: Hacker News Title: Show HN: Value likelihoods for OpenAI structured output Feedly Summary: Comments AI Summary and Description: Yes Summary: The text discusses the open-source Python library “structured-logprobs,” which enhances the understanding and reliability of outputs from OpenAI’s Language Learning Models (LLM) by providing detailed log probability information. This offers valuable…