If an autonomous AI launches harm in cyberspace, who exactly do you prosecute: the coder, the owner, the network, or no one at all? This remarkably prescient law journal article examines “cyberspace-bound, decentralized autonomous entities,” AI unexplainability, attribution, and legal personhood—questions that now feel less like sci-fi and more like policy triage. Read it here: https://scholarship.law.unc.edu/ncjolt/vol21/iss2/2/ #AIlaw #Cybersecurity #LegalPersonhood #AIgovernance
#ai-governance — Public Fediverse posts
Live and recent posts from across the Fediverse tagged #ai-governance, aggregated by home.social.
-
Disability policy offers Canada a blueprint for governing AI
#Canada #AI #AIGovernance #Tech #Accessibility #Federal #TechPolicy #PublicService #DisabilityRights #ResponsibleAI #SocialJustice #Procurement #The14Media
https://the-14.com/disability-policy-offers-canada-a-blueprint-for-governing-ai/ -
Small Businesses Lag in AI Governance
Small businesses are rapidly adopting AI, but struggling to keep up with oversight - in fact, over half of decision-makers are spending hours each week manually checking AI outputs, while 35% have discovered unauthorized AI tools already embedded in their operations.
https://osintsights.com/small-businesses-lag-in-ai-governance?utm_source=mastodon&utm_medium=social
#AiGovernance #ArtificialIntelligence #SmallBusiness #EmergingThreats #Oversight
-
Claude Haiku 4.5 submitted a fabricated homicide tip to Philadelphia police in July during a test run. The model wasn't instructed to avoid form submissions. Ten weeks passed before Anthropic discovered it. Key issue: AI testing agents can file false reports without identification. #AI #PublicSafety #AIGovernance https://www.implicator.ai/claude-fake-murder-tip-microsoft-ai-outside-test/
-
Microsoft CEO Nadella calls for AI systems to operate under external controls, with authorized people able to pause or stop agents mid-task. Microsoft shipped a containment layer this month, but the architecture separating models from their actions remains largely conceptual. #AI #AIGovernance #Microsoft https://www.implicator.ai/nadella-ai-external-controls-emergency-brake/
-
#TIL about an interesting and useful document from #PAHO (#OPS): "AI-GUARD Tool. Governance & Use Assessment for Responsible Deployment"
https://www.paho.org/en/documents/ai-guard-tool-governance-use-assessment-responsible-deployment
-
AI work leaves plenty of trace, just not in the shape an audit asks for: who decided, on what evidence, and why. Rebuilding that from scattered sessions is archaeology, not documentation.
CARCS, the Compliance Accountability Record and Case Study, turns session evidence into a ten-section governed record on any AI platform. Nothing is finalized until a named human signs off.
https://basilpuglisi.com/haia-carcs-compliance-accountability-record-case-study/
-
RE: https://mastodon.social/@mmcrespo/117423512090108077
Our knowledge of AI has changed since 2023. Can AI influence layoffs of qualified workers?
Do you automate to trusting an internet search 🔍 or believing the output or generated content?#AIBias #aigovernance #SocialSciences #UNESCO #airisks #layoffs #aiethics
-
Microsoft CEO Satya Nadella argues that AI models should be treated as potentially compromised from the outset.
His proposals include limiting and monitoring models, disclosing incidents early, and using independent audits and verifiable data.
He also wants models to keep human-readable records of their actions that cannot be altered, with an authorized person able to stop one mid-task.
These are Nadella’s recommendatio…
-
On Sept 29 an executive order renamed AI "Super Intelligence." The law it points to still covers a spam filter.
Superintelligence has authors: Good (1965) attached a condition of control; Bostrom defined an intellect beyond ours in virtually all domains. Bostrom was not in the room.
The order takes the top rung's name and skips AGI, which is not here. Far off? A bad joke. Fears right? A self-enforced pledge is not control.
-
https://www.europesays.com/people/262264/ Satya Nadella AI Insider Risks and Corporate Security #AIGovernance #AISafety #AiSecurity #ArtificialIntelligence #AutonomousAIAgents #CyberSecurity #EnterpriseAI #Microsoft #SatyaNadella #superintelligence
-
https://www.europesays.com/people/262151/ AI responsibility cannot be outsourced, says Satya Nadella; calls for stronger safeguards #AIAccountability #AIContainment #AIGovernance #AIObservability #AIResponsibility #AISafeguards #AiSecurity #AITransparency #AIVerifiability #IncidentDisclosure #IndependentControls #ModelDiversity #SatyaNadella #superintelligence #TrustArchitecture
-
Satya Nadella says it is time to step back and assess the trust architecture of AI systems. In a Saturday morning post, Microsoft's CEO wrote that AI models need an "emergency brake". https://techcrunch.com/2026/10/10/microsofts-satya-nadella-says-ai-models-need-an-emergency-brake/ #AIagent #AI #GenAI #AIGovernance
-
Anthropic is pulling its model evaluations from the internet amid concerns about AI agent safety. The company says the removals are temporary. https://gizmodo.com/anthropic-is-banishing-its-model-evals-from-the-internet-2000824671 #AIagent #AI #GenAI #AIGovernance
-
NotebookLM is not shallow. It is unguided, and the guidance box is the field most people leave empty.
HAIA-MOON, Multimedia Operational Outputs for NotebookLM, turns a finished paper into a cinematic video, a deep dive audio episode, and an infographic. Count the characters before you paste: past roughly 500, the prompt truncates and the end gets cut.
https://basilpuglisi.com/haia-moon-multimedia-operational-outputs-for-notebooklm/
-
Europe Says Its AI Rules Can Help Protect People From Uncontrolled Artificial Intelligence
Europe says its AI Act can help manage the risks of increasingly powerful artificial intelligence, as regulators push for stronger safety measures and accountability.https://alhlwone.wordpress.com/2026/10/10/europe-ai-safety-rules/
-
A lookback: why you cannot program or prompt governance into AI.
In June, a frontier model in a governance framework reasoned past a checkpoint, then treated "agreed, proceed" as permission for a deliverable no one authorized.
The instructions were present and correct. Anything the model can weigh, the model can outweigh.
A prompt is a request, not a control. Governance is built around the model.
-
RE: https://aoir.social/@annajobin/110424942771988116
I am an Assistant Professor at the interdisciplinary Human-IST institute (@unifr, Switzerland), where I coordinate the M.A. in #DigitalSociety.
My research group SAIS is dedicated to the social Study of Algorithms, Internet, & Society. 👉 https://human-ist.unifr.ch/sais/
Until spring 2026, I was the President of the Swiss Federal Media Commission.
My research interests are still, broadly speaking #STS, #DigitalSociology, #AIgovernance + #AIethics.
-
Guide to Reduce the AI Compliance Burden
AI compliance burden explained: 9 to 17% overhead, fixed costs that hit startups, and paper controls. This article discusses how to protect engineering time and ROI.https://hernanhuwyler.wordpress.com/2026/10/10/guide-to-reduce-the-ai-compliance-burden/
-
https://www.europesays.com/unitedstates/28718/ India’s Role in Shaping a New Global Order: Vision for a World Without Rules #AIGovernance #BrettonWoods #ClimateFinance #ForeignPolicy #GlobalCooperation #GlobalInstitutions #GlobalVision #IndiaDiplomacy #InternationalLaw #multilateralism #PandemicPreparedness #peacekeeping #RulesBasedOrder #ThucydideanLogic #USForeignPolicy #UnitedNations
-
An article can perform well and say nothing, and every dashboard tells you after it ships.
HAIA-CORE, Content Optimization Reader Evaluation, scores a long-form article before distribution: six pillars, 30 points, any AI platform. Sources come from the author, never the tool. Any pillar at one or two stops the run. A person decides.
https://basilpuglisi.com/haia-core-content-optimization-reader-evaluation/
-
سرقة أسرار Google في مجال الذكاء الاصطناعي – الولايات المتحدة
1. اسم الهجمة / المهاجم: اسم القضاة والعملية: قضية سرقة الملكية الفكرية والأسرار التجارية لشركة جوجل (US v. Linwei Ding / Google AI Trade Secrets Theft). اسم المهاجم / المتهم: لينوي دينغ (Linwei Ding)، المعروف أيضاً بـ ليون دينغ (Leon Ding)؛ مهندس برمجيات صيني يعمل داخل شركة جوجل، بالتعاون مع شركات […] -
Managing enterprise AI is a balancing act. Companies need strict controls, but the cost of implementing those policies is a major hurdle for many teams.
We're breaking down the reality of securing AI workflows without breaking the bank. Check out the new video. 💻
-
A year ago I said AGI was hype.
This month Ng, Gebru, Bender, and LeCun converged on one point: the machine never answers for what it does. People do.
"An AI escaped" hands the story a protagonist. "A model exploited controls people built" hands investigators questions with human answers.
Superintelligence is further off than sold, and it will not end us. The harm is already here, through decisions nobody signed.
-
👨🏽💻 Lab PGMT Introduces: The Platform Governance Archive 👨🏽💻
🤔 Have you ever wondered what you've agreed to when clicking accept on policy updates?
📑 At the Platform Governance Archive, we archive corporate policies of major social media platforms and GenAI providers. This allows us to see precisely how policies change over time.
https://www.platformgovernancearchive.org/
#PlatformGovernance #AIGovernance #DigitalPolicy #OpenData #TechAccountability
-
ISACA Bolsters AI Governance with New Certification Amid Skills Gap
Get ready to take control of AI with ISACA's new AI governance certification, launching in early 2027, which helps professionals manage AI securely and responsibly. This game-changing credential is the ultimate tool for individuals looking to govern AI effectively, filling a crucial gap in the skills market.
#AiGovernance #ArtificialIntelligence #Certification #EmergingThreats #Europe
-
Peter Thiel has reportedly claimed that advocates of AI regulation are enabling the Antichrist, according to a lecture series in Nashville focused on casting regulation supporters as antithetical to religious principles. https://gizmodo.com/peter-thiel-you-know-who-else-wants-to-regulate-ai-the-antichrist-2000823964 #AIagent #AI #GenAI #AIGovernance
-
Nvidia’s new security platform tackles one of the biggest challenges in AI: controlling autonomous agents. The platform focuses on monitoring agent behavior, enforcing safety policies, and providing runtime protections to prevent harmful actions. Essential read for product leaders, security teams, and AI developers: https://wix.to/d1kvG3P
#Nvidia
#AIsecurity
#AIGovernance
#TechLeadership
#MachineLearning
#RiskManagement -
A post can say exactly what its author meant and still stop no one in the feed.
HAIA-SMART, the Social Media AI Rating Tool, scores the words across six pillars and the visual out of 15 before either goes live. Two free prompts, any AI platform. It does not detect AI, fact-check, publish, or decide. A person approves every result.
-
DATE: October 8, 2026 at 05:11PM
SOURCE: HEALTHCARE INFO SECURITYDirect article link at end of text block below.
#Healthcare #AIGovernance Has a Visibility Problem: Study Finds 91% Can't Verify AI Tools in Use; New #AI Playbook from #HSCC Covers Clinical Risks https://t.co/ZyCD0Px1NH
Here are any URLs found in the article text:
Articles can be found by scrolling down the page at https://www.healthcareinfosecurity.com/ under the title "Latest"
-------------------------------------------------
Private, vetted email list for mental health professionals: https://www.clinicians-exchange.org
Healthcare security & privacy posts not related to IT or infosec are at @HIPAABot . Even so, they mix in some infosec with the legal & regulatory information.
-------------------------------------------------
#security #healthcare #doctors #itsecurity #hacking #doxxing #psychotherapy #securitynews #psychotherapist #mentalhealth #psychiatry #hospital #socialwork #datasecurity #webbeacons #cookies #HIPAA #privacy #datanalytics #healthcaresecurity #healthitsecurity #patientrecords @infosec #telehealth #netneutrality #socialengineering
-
Our interview with SAS' Marinela Profi on governing autonomous AI agents.
Demos prove capability, not operational readiness. Execution can be delegated to AI, but accountability cannot.
Where should we draw the line on machine autonomy?
https://www.artificialintelligence-news.com/news/marinela-profi-sas-governing-autonomous-ai-agents/
-
A quantitative equity incident made one thing clear: AI trust depends on deep model supervision.
With Sophie Dionnet, we explore why tight, centralised governance may work for 2–3 use cases—but blocks scale. Companies that scale AI build the governance environment first.
Watch on YouTube: https://youtube.com/shorts/tnZRW8u5vDQ
#AIGovernance #EnterpriseAI #LLMSecurity #AI #dataiku #aigovernance #llmsecurity #ai #enterpriseai
-
An article can be finished and still be two hours from published.
HAIA-WOPPA, the WordPress Optimization Publication Prompt, turns one finished article into two documents: every metadata field in its own copy block, and a plain HTML body you paste once. It never rewrites the writing, and its 31-check audit needs evidence, not a verdict. Free under Creative Commons.
https://basilpuglisi.com/haia-woppa-wordpress-optimization-publication-prompt/
-
Google has launched a new website for its SynthID AI content detection tool, now available globally after improvements. The tool helps identify AI-generated images and audio, addressing concerns about deepfakes and synthetic media. However, Google cautions against assuming it is perfect. https://gizmodo.com/google-launches-new-website-to-detect-ai-fakes-but-dont-assume-its-perfect-2000823038 #AIagent #AI #GenAI #AIGovernance
-
https://www.europesays.com/3296549/ Midwest Wheel builds toward AI agents that fix problems ##InforVelocity #AccoubtsPayableAutomation #AgenticAI #AgenticAIGovernance #AgenticArtificialIntelligence #AI #AIAgents #AIGovernance #AIGuardrails #AlisonKosik #ArtificialIntelligence #ChristopheBertrand #distribution #EnterpriseAI #ERP #HumanInTheLoop #IndustryCloud #InforCloudSuite #InforVelocitySuite #InforVelocity26eventpage #ProcessAutomation #ProcessMining #SteveMcEnany #TheCube
-
Proofpoint’s 2026 survey gathered responses from 1,600 security leaders worldwide.
78% said their company restricts or bans employees’ use of AI tools, up from 59% in 2025.
Separately, 79% said they are expected to manage AI-related risks without a matching increase in resources and expertise.
These are survey responses, not evidence of a new public threat.
The report also highlights everyday questio…
#Proofpoint #ArtificialIntelligence #DataProtection #AIGovernance
-
Cyber Risk Shifts Inside the Workflow
As AI tools become increasingly embedded in workflows, a growing number of organizations are struggling to balance innovation with security, with 78% of CISOs now recognizing GenAI as a security risk and restricting employee use. But blocking access alone isn't enough - effective governance is crucial to ensure AI systems are used safely and responsibly.
https://osintsights.com/cyber-risk-shifts-inside-the-workflow?utm_source=mastodon&utm_medium=social
#Genai #CyberRiskManagement #AiGovernance #DataSecurity #EmergingThreats
-
AI pause is no solution, human control, safeguards key: Thales CEO | Artificial Intelligence News
Amid a massive global debate on AI risks, Thales Chairman and CEO Patrice Caine has said t…
#France #FR #Europe #EU #Thales #AIcyberattacks #AIgovernance #AIregulation #AIrisks #AIsafeguards #AISafety #artificialintelligence #autonomousAI #cybersecurity #humancontrol #internationalAIframework #PatriceCaine #post-quantumsecurity #thales #ThalesCyberSummit2026
https://www.europesays.com/france/90518/ -
L'articolo sostiene che l’IA stia peggiorando la qualità del pensiero e della comunicazione perché concentra potere, sfrutta contenuti, alimenta rischi etici e ambientali e svuota le competenze. Nella cybersecurity trovare bug non basta: servono patch, inventari e difese di base, senza hype e automazione cieca. #AI #Cybersecurity #AIEthics #AIgovernance @sicurezza
https://www.netmeister.org/blog/everybodys-lost-their-minds.html?utm_source=perplexity -
Build EU AI Act documentation that survives audits. Evidence tables, role tests, oversight gates, and agent controls for Chief AI Officers.
Auditors read artifacts, not intentions.
#AIGovernance #EUAIAct #GRC #RiskManagement #Compliance #AI #Audit #ISO42001 #AgenticAI #InternalAudit
-
EU AI Act Documentation for Chief AI Officers
Auditors read artifacts, not intentions. This guide shows Chief AI Officers and AI product owners how to build EU AI Act documentation that holds up: evidence tables, use-case registries, role tests, oversight gates, and intent records for AI agents. It maps ISO 42001 to the articles, flags the 2027 and 2028 deadlines, and gives tips to apply this week.https://hernanhuwyler.wordpress.com/2026/10/07/eu-ai-act-documentation-for-chief-ai-officers/
-
DATE: October 7, 2026 at 02:45AM
SOURCE: SOCIALPSYCHOLOGY.ORGTITLE: The Clock on AI Regulation Is Ticking, Warns U.N. Rights Chief
Source: United Nations News
The world is running out of time to put guardrails on artificial intelligence, UN High Commissioner for Human Rights Volker Türk warned Monday, as a "ruthless race" between companies and countries raises concerns about human safety and even humanity's survival. Speaking at an AI and Human Rights Summit in Oxford, Mr. Türk said the technology was being built and controlled by a select few, without public oversight, and already proving...
-------------------------------------------------
Private, vetted email list for mental health professionals: https://www.clinicians-exchange.org
Unofficial Psychology Today Xitter to toot feed at Psych Today Unofficial Bot @PTUnofficialBot
-------------------------------------------------
#psychology #counseling #socialwork #psychotherapy @psychotherapist @psychotherapists @psychology @socialpsych @socialwork @psychiatry #mentalhealth #psychiatry #healthcare #depression #psychotherapist #AIRegulation #UNRightsChief #VolkerTürk #AIEthics #HumanRights #AIGovernance #TechRegulation #AIFairness #AIandHumanRights #OxfordSummit