<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/">
  <channel>
    <title>Safety on Tenu Tech Brief</title>
    <link>https://cluster-site.onrender.com/tags/safety/</link>
    <description>Recent content in Safety on Tenu Tech Brief</description>
    <generator>Hugo -- 0.146.0</generator>
    <language>en-us</language>
    <lastBuildDate>Wed, 25 Feb 2026 23:38:28 +0000</lastBuildDate>
    <atom:link href="https://cluster-site.onrender.com/tags/safety/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>NASA&#39;s Aerospace Safety Advisory Panel Releases 2025 Annual Report</title>
      <link>https://cluster-site.onrender.com/posts/nasas-aerospace-safety-advisory-panel-releases-2025-annual-report/</link>
      <pubDate>Wed, 25 Feb 2026 22:27:05 +0000</pubDate>
      <guid>https://cluster-site.onrender.com/posts/nasas-aerospace-safety-advisory-panel-releases-2025-annual-report/</guid>
      <description>• 5 min read NASA&amp;rsquo;s Aerospace Safety Advisory Panel Releases 2025 Annual Report Jessica Taveau NASA Headquarters The Aerospace Safety Advisory Panel (ASAP), which advises NASA and</description>
    </item>
    <item>
      <title>Canadian government demands safety changes from OpenAI</title>
      <link>https://cluster-site.onrender.com/posts/canadian-government-demands-safety-changes-from-openai/</link>
      <pubDate>Wed, 25 Feb 2026 20:49:24 +0000</pubDate>
      <guid>https://cluster-site.onrender.com/posts/canadian-government-demands-safety-changes-from-openai/</guid>
      <description>• Trending: Everything announced at Samsung Unpacked: New Galaxy S26 phones and more Alexa+ gets three new personality styles Apple event on March 4: What to expect Hands-on with G</description>
    </item>
    <item>
      <title>Anthropic weakens its safety pledge in the wake of the Pentagon&#39;s pressure campaign</title>
      <link>https://cluster-site.onrender.com/posts/anthropic-weakens-its-safety-pledge-in-the-wake-of-the-pentagons-pressure-campaign/</link>
      <pubDate>Wed, 25 Feb 2026 18:34:36 +0000</pubDate>
      <guid>https://cluster-site.onrender.com/posts/anthropic-weakens-its-safety-pledge-in-the-wake-of-the-pentagons-pressure-campaign/</guid>
      <description>• Trending: Everything announced at Samsung Unpacked: New Galaxy S26 phones and more Samsung Galaxy Unpacked 2026: Live updates on Galaxy S26 launch Alexa+ gets three new personali</description>
    </item>
    <item>
      <title>When can we trust untrusted monitoring? A safety case sketch across collusion strategies</title>
      <link>https://cluster-site.onrender.com/posts/when-can-we-trust-untrusted-monitoring-a-safety-case-sketch-across-collusion-strategies/</link>
      <pubDate>Wed, 25 Feb 2026 05:00:00 +0000</pubDate>
      <guid>https://cluster-site.onrender.com/posts/when-can-we-trust-untrusted-monitoring-a-safety-case-sketch-across-collusion-strategies/</guid>
      <description>• Computer Science &amp;gt; Artificial Intelligence [Submitted on 24 Feb 2026] Title:When can we trust untrusted monitoring? • A safety case sketch across collusion strategies View PDF HT</description>
    </item>
    <item>
      <title>Reddit, porn sites fined by UK regulators over children&#39;s safety and privacy</title>
      <link>https://cluster-site.onrender.com/posts/reddit-porn-sites-fined-by-uk-regulators-over-childrens-safety-and-privacy/</link>
      <pubDate>Tue, 24 Feb 2026 15:48:02 +0000</pubDate>
      <guid>https://cluster-site.onrender.com/posts/reddit-porn-sites-fined-by-uk-regulators-over-childrens-safety-and-privacy/</guid>
      <description>• Reddit, porn sites fined by UK regulators over children&amp;rsquo;s safety and privacy The UK&amp;rsquo;s online safety and privacy regulators are targeting companies that violate new age verificati</description>
    </item>
    <item>
      <title>Northeast Blizzard Delivers Nearly 3 Feet of Snow Across Region</title>
      <link>https://cluster-site.onrender.com/posts/northeast-blizzard-delivers-nearly-3-feet-of-snow-across-region/</link>
      <pubDate>Tue, 24 Feb 2026 07:07:54 +0000</pubDate>
      <guid>https://cluster-site.onrender.com/posts/northeast-blizzard-delivers-nearly-3-feet-of-snow-across-region/</guid>
      <description>• Northeast blizzard delivered nearly 3 feet of snow across the region. • Storm classified as strongest in a decade and among most intense historically. • Meteorologists noted powe</description>
    </item>
    <item>
      <title>How to Restore your 19th-Century Lancashire Boiler to Hold 120 PSI</title>
      <link>https://cluster-site.onrender.com/posts/how-to-restore-your-19th-century-lancashire-boiler-to-hold-120-psi/</link>
      <pubDate>Tue, 24 Feb 2026 03:00:33 +0000</pubDate>
      <guid>https://cluster-site.onrender.com/posts/how-to-restore-your-19th-century-lancashire-boiler-to-hold-120-psi/</guid>
      <description>• Lancashire boilers from 1930s now refurbished at Claymills Pumping Station. • Boiler 1 last used 1971, now restored to 120 PSI safety margin. • Volunteers removed decades of scal</description>
    </item>
    <item>
      <title>Nissan&#39;s &#39;moving battery pack&#39; patent aims to boost EV safety, performance</title>
      <link>https://cluster-site.onrender.com/posts/nissans-moving-battery-pack-patent-aims-to-boost-ev-safety-performance/</link>
      <pubDate>Mon, 23 Feb 2026 13:12:18 +0000</pubDate>
      <guid>https://cluster-site.onrender.com/posts/nissans-moving-battery-pack-patent-aims-to-boost-ev-safety-performance/</guid>
      <description>• Nissan&amp;rsquo;s &amp;lsquo;moving battery pack&amp;rsquo; patent aims to boost EV safety, performance The system will use actuators controlled by a central computer to decide optimal position of the batter</description>
    </item>
    <item>
      <title>Meet Regina Senegal, Acting Chief of Johnson&#39;s Quality and Flight Equipment Division</title>
      <link>https://cluster-site.onrender.com/posts/meet-regina-senegal-acting-chief-of-johnsons-quality-and-flight-equipment-division/</link>
      <pubDate>Mon, 23 Feb 2026 10:00:00 +0000</pubDate>
      <guid>https://cluster-site.onrender.com/posts/meet-regina-senegal-acting-chief-of-johnsons-quality-and-flight-equipment-division/</guid>
      <description>• Regina Senegal leads safety and quality teams for 13 NASA customers, including Orion and Gateway. • Her division supports program-level requirement writing, assurance programs, a</description>
    </item>
    <item>
      <title>Toyota may block drivers from disabling safety features on its cars</title>
      <link>https://cluster-site.onrender.com/posts/toyota-may-block-drivers-from-disabling-safety-features-on-its-cars/</link>
      <pubDate>Sat, 21 Feb 2026 05:03:00 +0000</pubDate>
      <guid>https://cluster-site.onrender.com/posts/toyota-may-block-drivers-from-disabling-safety-features-on-its-cars/</guid>
      <description>• Toyota may block drivers from disabling safety features on its cars Facebook Twitter LinkedIn Reddit Flipboard Email Times they are a-changin, and what was once deemed as a set o</description>
    </item>
    <item>
      <title>Ghost Robotics: Innovating for safety</title>
      <link>https://cluster-site.onrender.com/posts/ghost-robotics-innovating-for-safety/</link>
      <pubDate>Sat, 21 Feb 2026 00:40:33 +0000</pubDate>
      <guid>https://cluster-site.onrender.com/posts/ghost-robotics-innovating-for-safety/</guid>
      <description>• In Episode 232 ofThe Robot ReportPodcast, Gavin Kenneally, co-founder and CEO of Ghost Robotics, discusses the journey of his company, from its inception to becoming a leader in</description>
    </item>
    <item>
      <title>Intent Laundering: AI Safety Datasets Are Not What They Seem</title>
      <link>https://cluster-site.onrender.com/posts/intent-laundering-ai-safety-datasets-are-not-what-they-seem/</link>
      <pubDate>Fri, 20 Feb 2026 05:00:00 +0000</pubDate>
      <guid>https://cluster-site.onrender.com/posts/intent-laundering-ai-safety-datasets-are-not-what-they-seem/</guid>
      <description>• Computer Science &amp;gt; Cryptography and Security [Submitted on 17 Feb 2026] Title:Intent Laundering: AI Safety Datasets Are Not What They Seem View PDF HTML (experimental)Abstract:We</description>
    </item>
    <item>
      <title>Mind the GAP: Text Safety Does Not Transfer to Tool-Call Safety in LLM Agents</title>
      <link>https://cluster-site.onrender.com/posts/mind-the-gap-text-safety-does-not-transfer-to-tool-call-safety-in-llm-agents/</link>
      <pubDate>Fri, 20 Feb 2026 05:00:00 +0000</pubDate>
      <guid>https://cluster-site.onrender.com/posts/mind-the-gap-text-safety-does-not-transfer-to-tool-call-safety-in-llm-agents/</guid>
      <description>• Computer Science &amp;gt; Artificial Intelligence [Submitted on 18 Feb 2026] Title:Mind the GAP: Text Safety Does Not Transfer to Tool-Call Safety in LLM Agents View PDF HTML (experimen</description>
    </item>
    <item>
      <title>Narrow fine-tuning erodes safety alignment in vision-language agents</title>
      <link>https://cluster-site.onrender.com/posts/narrow-fine-tuning-erodes-safety-alignment-in-vision-language-agents/</link>
      <pubDate>Fri, 20 Feb 2026 05:00:00 +0000</pubDate>
      <guid>https://cluster-site.onrender.com/posts/narrow-fine-tuning-erodes-safety-alignment-in-vision-language-agents/</guid>
      <description>• Computer Science &amp;gt; Artificial Intelligence [Submitted on 18 Feb 2026] Title:Narrow fine-tuning erodes safety alignment in vision-language agents View PDF HTML (experimental)Abstr</description>
    </item>
    <item>
      <title>ICYMI: Memory Safety, Ecosystem Talks, and Java Interoperability at FOSDEM 2025</title>
      <link>https://cluster-site.onrender.com/posts/icymi-memory-safety-ecosystem-talks-and-java-interoperability-at-fosdem-2025/</link>
      <pubDate>Fri, 20 Feb 2026 00:37:50 +0000</pubDate>
      <guid>https://cluster-site.onrender.com/posts/icymi-memory-safety-ecosystem-talks-and-java-interoperability-at-fosdem-2025/</guid>
      <description>• ICYMI: Memory Safety, Ecosystem Talks, and Java Interoperability at FOSDEM 2025 The Swift community had a strong presence at FOSDEM 2025, the world&amp;rsquo;s largest independently run op</description>
    </item>
    <item>
      <title>AI Agents Are Getting Better. Their Safety Disclosures Aren&#39;t</title>
      <link>https://cluster-site.onrender.com/posts/ai-agents-are-getting-better.-their-safety-disclosures-arent/</link>
      <pubDate>Thu, 19 Feb 2026 23:43:07 +0000</pubDate>
      <guid>https://cluster-site.onrender.com/posts/ai-agents-are-getting-better.-their-safety-disclosures-arent/</guid>
      <description>• ai atlas Tech Services &amp;amp; Software AI AI Agents Are Getting Better. • Their Safety Disclosures Aren&amp;rsquo;t A study led by MIT researchers found that agentic AI developers seldom publis</description>
    </item>
    <item>
      <title>US Air Force aims to enhance aircraft safety with improved data management</title>
      <link>https://cluster-site.onrender.com/posts/us-air-force-aims-to-enhance-aircraft-safety-with-improved-data-management/</link>
      <pubDate>Thu, 19 Feb 2026 21:44:27 +0000</pubDate>
      <guid>https://cluster-site.onrender.com/posts/us-air-force-aims-to-enhance-aircraft-safety-with-improved-data-management/</guid>
      <description>• US Air Force aims to enhance aircraft safety with improved data management The US Air Force will utilize a new data management method to enhance aircraft safety and reduce mainte</description>
    </item>
    <item>
      <title>Official News regarding the CERN Safety Rules: Oxygen Deficiency Hazard</title>
      <link>https://cluster-site.onrender.com/posts/official-news-regarding-the-cern-safety-rules-oxygen-deficiency-hazard/</link>
      <pubDate>Thu, 19 Feb 2026 09:02:56 +0000</pubDate>
      <guid>https://cluster-site.onrender.com/posts/official-news-regarding-the-cern-safety-rules-oxygen-deficiency-hazard/</guid>
      <description>• Voir en Official News regarding the CERN Safety Rules: Oxygen Deficiency Hazard 19 February, 2026 | ByHSE unit The CERN Safety Rules listed below have been published on the CERN</description>
    </item>
    <item>
      <title>ICYMI: Memory Safety, Ecosystem Talks, and Java Interoperability at FOSDEM 2025</title>
      <link>https://cluster-site.onrender.com/posts/icymi-memory-safety-ecosystem-talks-and-java-interoperability-at-fosdem-2025/</link>
      <pubDate>Thu, 19 Feb 2026 01:56:11 +0000</pubDate>
      <guid>https://cluster-site.onrender.com/posts/icymi-memory-safety-ecosystem-talks-and-java-interoperability-at-fosdem-2025/</guid>
      <description>• ICYMI: Memory Safety, Ecosystem Talks, and Java Interoperability at FOSDEM 2025 The Swift community had a strong presence at FOSDEM 2025, the world&amp;rsquo;s largest independently run op</description>
    </item>
    <item>
      <title>Social media on trial: tech giants face lawsuits over addiction, safety, and mental health</title>
      <link>https://cluster-site.onrender.com/posts/social-media-on-trial-tech-giants-face-lawsuits-over-addiction-safety-and-mental-health/</link>
      <pubDate>Wed, 18 Feb 2026 21:27:09 +0000</pubDate>
      <guid>https://cluster-site.onrender.com/posts/social-media-on-trial-tech-giants-face-lawsuits-over-addiction-safety-and-mental-health/</guid>
      <description>• Policy Social media on trial: tech giants face lawsuits over addiction, safety, and mental health Link Share RSS Gift A set of bellwether cases alleging that social media platfor</description>
    </item>
    <item>
      <title>China unveils Dragon Armor 3.0 battery with world-first fire-energy separation design</title>
      <link>https://cluster-site.onrender.com/posts/china-unveils-dragon-armor-3.0-battery-with-world-first-fire-energy-separation-design/</link>
      <pubDate>Wed, 18 Feb 2026 17:32:51 +0000</pubDate>
      <guid>https://cluster-site.onrender.com/posts/china-unveils-dragon-armor-3.0-battery-with-world-first-fire-energy-separation-design/</guid>
      <description>• Dragon Armor 3.0 introduces world‑first fire‑energy separation in Li‑ion batteries. • Physical isolation of terminals from pressure relief channels reduces thermal runaway risk.</description>
    </item>
    <item>
      <title>CEOs who experience natural disasters are more likely to lead safer workplaces</title>
      <link>https://cluster-site.onrender.com/posts/ceos-who-experience-natural-disasters-are-more-likely-to-lead-safer-workplaces/</link>
      <pubDate>Wed, 18 Feb 2026 17:25:27 +0000</pubDate>
      <guid>https://cluster-site.onrender.com/posts/ceos-who-experience-natural-disasters-are-more-likely-to-lead-safer-workplaces/</guid>
      <description>• CEOs who survived natural disasters prioritize workplace safety, reducing injury rates. • Their firsthand crisis experience translates into stronger safety protocols for employee</description>
    </item>
    <item>
      <title>Cracks in snow propagate faster than expected</title>
      <link>https://cluster-site.onrender.com/posts/cracks-in-snow-propagate-faster-than-expected/</link>
      <pubDate>Wed, 18 Feb 2026 14:40:04 +0000</pubDate>
      <guid>https://cluster-site.onrender.com/posts/cracks-in-snow-propagate-faster-than-expected/</guid>
      <description>• Hundreds of &amp;lsquo;whumpfs&amp;rsquo; reported since Jan 10, 2026, indicating widespread weak snow layers. • Whumpfs are sudden fractures that can quickly spread across terrain, risking avalanch</description>
    </item>
    <item>
      <title>Cultured beef differs from conventional beef in allergy-related hazards, food safety study shows</title>
      <link>https://cluster-site.onrender.com/posts/cultured-beef-differs-from-conventional-beef-in-allergy-related-hazards-food-safety-study-shows/</link>
      <pubDate>Wed, 18 Feb 2026 13:00:01 +0000</pubDate>
      <guid>https://cluster-site.onrender.com/posts/cultured-beef-differs-from-conventional-beef-in-allergy-related-hazards-food-safety-study-shows/</guid>
      <description>• As cultured meat moves toward commercialization, people want to understand how it impacts health compared to conventional animal meat. • So, researchers publishing in theJournal</description>
    </item>
    <item>
      <title>Video: World&#39;s largest electric vehicle reaches crucial safety milestone before launch</title>
      <link>https://cluster-site.onrender.com/posts/video-worlds-largest-electric-vehicle-reaches-crucial-safety-milestone-before-launch/</link>
      <pubDate>Tue, 17 Feb 2026 11:07:46 +0000</pubDate>
      <guid>https://cluster-site.onrender.com/posts/video-worlds-largest-electric-vehicle-reaches-crucial-safety-milestone-before-launch/</guid>
      <description>• Video: World&amp;rsquo;s largest electric vehicle reaches crucial safety milestone before launch The ship carries 250 tons of batteries producing 40MW, equal to 487 Teslas and 4x larger th</description>
    </item>
    <item>
      <title>Performing an Autopsy on 15 Dead Battle Born LFP Batteries</title>
      <link>https://cluster-site.onrender.com/posts/performing-an-autopsy-on-15-dead-battle-born-lfp-batteries/</link>
      <pubDate>Tue, 17 Feb 2026 06:00:51 +0000</pubDate>
      <guid>https://cluster-site.onrender.com/posts/performing-an-autopsy-on-15-dead-battle-born-lfp-batteries/</guid>
      <description>• 15 dead Battle Born LFP batteries examined, expanding dataset beyond Will Prowse&amp;rsquo;s initial samples. • Symptoms varied: charging refusal, overheating terminals, no life, charged c</description>
    </item>
    <item>
      <title>ProMoral-Bench: Evaluating Prompting Strategies for Moral Reasoning and Safety in LLMs</title>
      <link>https://cluster-site.onrender.com/posts/promoral-bench-evaluating-prompting-strategies-for-moral-reasoning-and-safety-in-llms/</link>
      <pubDate>Tue, 17 Feb 2026 05:00:00 +0000</pubDate>
      <guid>https://cluster-site.onrender.com/posts/promoral-bench-evaluating-prompting-strategies-for-moral-reasoning-and-safety-in-llms/</guid>
      <description>• Computer Science &amp;gt; Artificial Intelligence [Submitted on 5 Feb 2026] Title:ProMoral-Bench: Evaluating Prompting Strategies for Moral Reasoning and Safety in LLMs View PDF HTML (e</description>
    </item>
    <item>
      <title>ProMoral-Bench: Evaluating Prompting Strategies for Moral Reasoning and Safety in LLMs</title>
      <link>https://cluster-site.onrender.com/posts/promoral-bench-evaluating-prompting-strategies-for-moral-reasoning-and-safety-in-llms/</link>
      <pubDate>Tue, 17 Feb 2026 05:00:00 +0000</pubDate>
      <guid>https://cluster-site.onrender.com/posts/promoral-bench-evaluating-prompting-strategies-for-moral-reasoning-and-safety-in-llms/</guid>
      <description>• Computer Science &amp;gt; Artificial Intelligence [Submitted on 5 Feb 2026] Title:ProMoral-Bench: Evaluating Prompting Strategies for Moral Reasoning and Safety in LLMs View PDF HTML (e</description>
    </item>
    <item>
      <title>Stay in Character, Stay Safe: Dual-Cycle Adversarial Self-Evolution for Safety Role-Playing Agents</title>
      <link>https://cluster-site.onrender.com/posts/stay-in-character-stay-safe-dual-cycle-adversarial-self-evolution-for-safety-role-playing-agents/</link>
      <pubDate>Tue, 17 Feb 2026 05:00:00 +0000</pubDate>
      <guid>https://cluster-site.onrender.com/posts/stay-in-character-stay-safe-dual-cycle-adversarial-self-evolution-for-safety-role-playing-agents/</guid>
      <description>• Computer Science &amp;gt; Artificial Intelligence [Submitted on 29 Jan 2026] Title:Stay in Character, Stay Safe: Dual-Cycle Adversarial Self-Evolution for Safety Role-Playing Agents Vie</description>
    </item>
    <item>
      <title>Stay in Character, Stay Safe: Dual-Cycle Adversarial Self-Evolution for Safety Role-Playing Agents</title>
      <link>https://cluster-site.onrender.com/posts/stay-in-character-stay-safe-dual-cycle-adversarial-self-evolution-for-safety-role-playing-agents/</link>
      <pubDate>Tue, 17 Feb 2026 05:00:00 +0000</pubDate>
      <guid>https://cluster-site.onrender.com/posts/stay-in-character-stay-safe-dual-cycle-adversarial-self-evolution-for-safety-role-playing-agents/</guid>
      <description>• Computer Science &amp;gt; Artificial Intelligence [Submitted on 29 Jan 2026] Title:Stay in Character, Stay Safe: Dual-Cycle Adversarial Self-Evolution for Safety Role-Playing Agents Vie</description>
    </item>
    <item>
      <title>Safety review completed at South African research reactor</title>
      <link>https://cluster-site.onrender.com/posts/safety-review-completed-at-south-african-research-reactor/</link>
      <pubDate>Thu, 12 Feb 2026 11:07:28 +0000</pubDate>
      <guid>https://cluster-site.onrender.com/posts/safety-review-completed-at-south-african-research-reactor/</guid>
      <description>• Safety review completed at South African research reactor The five-day six-person Safety Review Mission on Ageing Management and Continued Safe Operation was at the invitation of</description>
    </item>
    <item>
      <title>A one-prompt attack that breaks LLM safety alignment</title>
      <link>https://cluster-site.onrender.com/posts/a-one-prompt-attack-that-breaks-llm-safety-alignment/</link>
      <pubDate>Mon, 09 Feb 2026 17:12:11 +0000</pubDate>
      <guid>https://cluster-site.onrender.com/posts/a-one-prompt-attack-that-breaks-llm-safety-alignment/</guid>
      <description>• Share Link copied to clipboard! • Content types Research Topics Actionable threat insights AI and agents Security management Large language models (LLMs) and diffusion models now</description>
    </item>
    <item>
      <title>Forget IPA for SLA print washing, Prusament Resin Cleaner is better in every way</title>
      <link>https://cluster-site.onrender.com/posts/forget-ipa-for-sla-print-washing-prusament-resin-cleaner-is-better-in-every-way/</link>
      <pubDate>Fri, 23 Jan 2026 10:00:04 +0000</pubDate>
      <guid>https://cluster-site.onrender.com/posts/forget-ipa-for-sla-print-washing-prusament-resin-cleaner-is-better-in-every-way/</guid>
      <description>• Prusament Resin Cleaner DPM offers a safer alternative to IPA for SLA print washing. • DPM is far less flammable, with a higher ignition temperature than IPA. • Lower volatility</description>
    </item>
    <item>
      <title>AssetOpsBench: Bridging the Gap Between AI Agent Benchmarks and Industrial Reality</title>
      <link>https://cluster-site.onrender.com/posts/assetopsbench-bridging-the-gap-between-ai-agent-benchmarks-and-industrial-reality/</link>
      <pubDate>Wed, 21 Jan 2026 06:25:31 +0000</pubDate>
      <guid>https://cluster-site.onrender.com/posts/assetopsbench-bridging-the-gap-between-ai-agent-benchmarks-and-industrial-reality/</guid>
      <description>• AssetOpsBench introduces a multi-agent benchmark for industrial asset lifecycle management. • Covers 2.3M sensor telemetry points and 4.2K work orders. • Evaluates agents across</description>
    </item>
    <item>
      <title>Advancing Our Chef Infrastructure: Safety Without Disruption</title>
      <link>https://cluster-site.onrender.com/posts/advancing-our-chef-infrastructure-safety-without-disruption/</link>
      <pubDate>Thu, 23 Oct 2025 18:17:56 +0000</pubDate>
      <guid>https://cluster-site.onrender.com/posts/advancing-our-chef-infrastructure-safety-without-disruption/</guid>
      <description>• Last year, I wrote a blog post titled Advancing Our Chef Infrastructure, where we explored the evolution of our Chef infrastructure over the years. • We talked about the shift fr</description>
    </item>
    <item>
      <title>Advancing Our Chef Infrastructure: Safety Without Disruption</title>
      <link>https://cluster-site.onrender.com/posts/advancing-our-chef-infrastructure-safety-without-disruption/</link>
      <pubDate>Thu, 23 Oct 2025 18:17:56 +0000</pubDate>
      <guid>https://cluster-site.onrender.com/posts/advancing-our-chef-infrastructure-safety-without-disruption/</guid>
      <description>• Last year, I wrote a blog post titled Advancing Our Chef Infrastructure, where we explored the evolution of our Chef infrastructure over the years. • We talked about the shift fr</description>
    </item>
    <item>
      <title>Deploy Safety: Reducing customer impact from change</title>
      <link>https://cluster-site.onrender.com/posts/deploy-safety-reducing-customer-impact-from-change/</link>
      <pubDate>Tue, 07 Oct 2025 16:33:12 +0000</pubDate>
      <guid>https://cluster-site.onrender.com/posts/deploy-safety-reducing-customer-impact-from-change/</guid>
      <description>• It&amp;rsquo;s mid 2023 and we&amp;rsquo;ve identified some opportunities to improve our reliability. • Fast forward to January 2025. • Customer impact hours are reduced from the peak by 90% and con</description>
    </item>
    <item>
      <title>Deploy Safety: Reducing customer impact from change</title>
      <link>https://cluster-site.onrender.com/posts/deploy-safety-reducing-customer-impact-from-change/</link>
      <pubDate>Tue, 07 Oct 2025 16:33:12 +0000</pubDate>
      <guid>https://cluster-site.onrender.com/posts/deploy-safety-reducing-customer-impact-from-change/</guid>
      <description>• It&amp;rsquo;s mid 2023 and we&amp;rsquo;ve identified some opportunities to improve our reliability. • Fast forward to January 2025. • Customer impact hours are reduced from the peak by 90% and con</description>
    </item>
    <item>
      <title>Beyond the Moat: Why There Is Safety in Layers</title>
      <link>https://cluster-site.onrender.com/posts/beyond-the-moat-why-there-is-safety-in-layers/</link>
      <pubDate>Fri, 29 Aug 2025 14:02:59 +0000</pubDate>
      <guid>https://cluster-site.onrender.com/posts/beyond-the-moat-why-there-is-safety-in-layers/</guid>
      <description>• IT Security Commentary Insight and analysis on the information technology space from industry thought leaders. • Beyond the Moat: Why There Is Safety in Layers From medieval cast</description>
    </item>
    <item>
      <title>How to Evaluate Jailbreak Methods: A Case Study with the StrongREJECT Benchmark</title>
      <link>https://cluster-site.onrender.com/posts/how-to-evaluate-jailbreak-methods-a-case-study-with-the-strongreject-benchmark/</link>
      <pubDate>Wed, 28 Aug 2024 15:30:00 +0000</pubDate>
      <guid>https://cluster-site.onrender.com/posts/how-to-evaluate-jailbreak-methods-a-case-study-with-the-strongreject-benchmark/</guid>
      <description>• Researchers tested jailbreak via Scots Gaelic translation, initially replicating 43% success claim. • GPT-4 responded with bomb instructions in Gaelic, but full output differed f</description>
    </item>
  </channel>
</rss>
