![\"Text-to-Speech](\"https://www.artificialintelligence-news.com/wp-content/uploads/2026/09/gemini-3.8-flash-tts-text-to-speech-benchmark-gemini-3.8-flash-google-ai.jpg\")
Google has launched two Gemini 3.8 Flash TTS voice models, introducing dedicated speech generation systems engineered for direct performance scripting and high-volume audio production. The dual release splits vocal synthesis tasks between creative direction and cost-managed infrastructure. Gemini 3.8 Flash TTS targets interactive entertainment, game development, and long-form narrations where studio teams demand prompt-based vocal […]
\nThe post Google launches Gemini 3.8 Flash TTS voice models appeared first on AI News.
","content_html":"Google has launched two Gemini 3.8 Flash TTS voice models, introducing dedicated speech generation systems engineered for direct performance scripting and high-volume audio production. The dual release splits vocal synthesis tasks between creative direction and cost-managed infrastructure.
\n\n\n\nGemini 3.8 Flash TTS targets interactive entertainment, game development, and long-form narrations where studio teams demand prompt-based vocal design. Gemini 3.8 Flash-Lite TTS focuses on automated media dubbing, customer-facing conversational agents, and high-throughput translation pipelines.
\n\n\n\nBoth systems expand Google’s established audio roster, which previously introduced 3.5 Live Translate, 3.5 Transcribe, 3.8 Live, and 3.8 Live Extended Thinking. Technical teams structured the new models to replace fixed catalogues of 30 legacy voices.
\n\n\n\nDevelopers can now tap into a directory containing more than 2,000 pre-built vocal profiles covering regional linguistic variations such as Quebec French, Scots English, and Mexican Spanish across more than 100 languages. A forthcoming voice remixing module will let audio engineers adjust timbre, pitch, pace, and accent contours through direct text commands.
\n\n\n\nIndependent evaluations place the larger model at the top of third-party audio rankings.
\n\n\n\nOn the Hume AI Voice Design Benchmark, Gemini 3.8 Flash TTS registered an overall score of 71.4, alongside a category-leading 60.8 rating in accent modelling.
\n\n\n\nIn the Hume AI Overall Quality Index, Gemini 3.8 Flash TTS captured the top position while Gemini 3.8 Flash-Lite TTS secured second place, outperforming earlier baselines established by Gemini 3.1 Flash TTS.
\n\n\n\nDouble-blind human trials conducted through Voice Arena confirmed preference advantages in regional languages, including Japanese, Brazilian Portuguese, Vietnamese, Modern Standard Arabic, Mexican Spanish, and Hindi.
\n\n\n\nMulti-speaker staging and extended synthesis runs is a key focus. Single scripts direct dual-speaker exchanges, which preserve natural conversational turn-taking and vocal separation across prolonged dialogues.
\n\n\n\nAudio quality and character timbre remain stable over multi-hour files, cutting vocal degradation during audiobook recordings and episodic podcasts.
\n\n\n\nWriters can insert non-verbal acoustic markers directly into production text, dropping tags such as <laughs>, <sigh>, or <gasp> into lines, alongside reactive verbal interjections like |mhm| and |yeah| to balance conversational tempo.
\n\n\n\nVoice cloning pipelines rely on mandatory identity checks to counter impersonation risks. Recreating a vocal profile requires a 30-second reference recording accompanied by an explicit verbal consent track spoken by the original voice owner. Google validates acoustic alignment between both audio tracks before processing custom profiles.
\n\n\n\nGenerated sound files embed imperceptible SynthID audio watermarks and cryptographic C2PA provenance metadata directly into the exported waveform, ensuring downstream detection tools can identify synthetic speech assets.
\n\n\n\nDeployment across enterprise environments has begun across multiple software ecosystems. Software developers can access both Flash TTS and Flash-Lite TTS through Google AI Studio and the standard Gemini API, connecting to developer frameworks run by Agora, LiveKit, Pipecat, and Vercel. Early commercial integrations span Figma, HeyGen, Linguana, Wondercraft, 99.co, and Ollang for regional media translation and customer service automation.
\n\n\n\nEnd users receive Flash TTS directly inside Gemini Notebook, whereas Google Vids incorporates Flash-Lite TTS. Gemini Enterprise customers will receive administrative API access in an upcoming deployment wave.
\n\n\n\nSee also: U.S. TRANSCOM deploys randomised AI to secure military logistics
\n\n\n\n![\"Banner](\"https://www.artificialintelligence-news.com/wp-content/uploads/2026/09/image.png\")
Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is part of TechEx and is co-located with other leading technology events including IoT Tech Expo and the Cyber Security & Cloud Expo. Click here for more information.
\n\n\n\nAI News is powered by TechForge Media. Explore other upcoming enterprise technology events and webinars here.
\nThe post Google launches Gemini 3.8 Flash TTS voice models appeared first on AI News.
","content_text":"Google has launched two\nGemini 3.8 Flash\nTTS voice models, introducing dedicated speech generation systems engineered for direct performance scripting and high-volume audio production. The dual release splits vocal synthesis tasks between creative direction and cost-managed infrastructure.\nGemini 3.8 Flash TTS targets interactive entertainment, game development, and long-form narrations where studio teams demand prompt-based vocal design. Gemini 3.8 Flash-Lite TTS focuses on automated media dubbing, customer-facing conversational agents, and high-throughput translation pipelines.\nBoth systems expand Google’s established audio roster, which previously introduced 3.5 Live Translate, 3.5 Transcribe, 3.8 Live, and 3.8 Live Extended Thinking. Technical teams structured the new models to replace fixed catalogues of 30 legacy voices.\nDevelopers can now tap into a directory containing more than 2,000 pre-built vocal profiles covering regional linguistic variations such as Quebec French, Scots English, and Mexican Spanish across more than 100 languages. A forthcoming voice remixing module will let audio engineers adjust timbre, pitch, pace, and accent contours through direct text commands.\nHume AI benchmarks test synthesis performance\nIndependent evaluations place the larger model at the top of third-party audio rankings.\nOn the Hume AI Voice Design Benchmark, Gemini 3.8 Flash TTS registered an overall score of 71.4, alongside a category-leading 60.8 rating in accent modelling.\nIn the Hume AI Overall Quality Index, Gemini 3.8 Flash TTS captured the top position while Gemini 3.8 Flash-Lite TTS secured second place, outperforming earlier baselines established by Gemini 3.1 Flash TTS.\nDouble-blind human trials conducted through Voice Arena confirmed preference advantages in regional languages, including Japanese, Brazilian Portuguese, Vietnamese, Modern Standard Arabic, Mexican Spanish, and Hindi.\nMulti-speaker staging and extended synthesis runs is a key focus. Single scripts direct dual-speaker exchanges, which preserve natural conversational turn-taking and vocal separation across prolonged dialogues.\nAudio quality and character timbre remain stable over multi-hour files, cutting vocal degradation during audiobook recordings and episodic podcasts.\nWriters can insert non-verbal acoustic markers directly into production text, dropping tags such asMulti-agent AI systems are taking over supply chain execution as enterprise networks face diminishing returns from static dashboards, pushing logistics directors towards autonomous execution. Predictive demand models display recommendations, yet human planners still clear every action. Multi-agent systems replace that approval stage across targeted operational boundaries. Instead of waiting for weekly scheduling runs, independent software […]
\nThe post Multi-agent AI systems are taking over supply chain execution appeared first on AI News.
","content_html":"Multi-agent AI systems are taking over supply chain execution as enterprise networks face diminishing returns from static dashboards, pushing logistics directors towards autonomous execution.
\n\n\n\nPredictive demand models display recommendations, yet human planners still clear every action. Multi-agent systems replace that approval stage across targeted operational boundaries.
\n\n\n\nInstead of waiting for weekly scheduling runs, independent software models ingest real-time telemetry from carrier ETAs, yard cameras, and warehouse management system events. The agents execute freight re-routing, safety stock rebalancing, and dock allocations directly inside enterprise resource software.
\n\n\n\nLenovo reported this operational transition on its global iChain infrastructure across 180 markets, more than 30 factories, and 100 logistics centres. The hardware manufacturer linked an Order Fulfilment Agent and a Risk Management Agent directly to existing transaction platforms.
\n\n\n\nAccording to Lenovo, fulfilment decisions ran three times faster, disruption response four times faster, risk assessment operated at 85 percent accuracy, and delivery accuracy increased 30 percent.
\n\n\n\nA mid-size automotive parts manufacturer, documented by Simor Consulting, deployed five specialised agents across 15 countries and 200 suppliers over an 18-month production run. The company recorded an on-time delivery rise from 82 percent to 94 percent.
\n\n\n\nThe manufacturer observed that its disruption agent detected supply threats 48 hours ahead of manual monitoring teams. Communication agents interacted smoothly with longstanding suppliers. Dialogue failed with unfamiliar vendors until the software catalogued their specific reply behaviours.
\n\n\n\nInter-enterprise logistics routing trials show comparable results. An initial virtual-network exercise conducted by Fujitsu and Rohto Pharmaceutical yielded transport cost reductions of up to 30 percent. Both companies scheduled an expanded trial on Rohto’s live physical chain between January 2026 and March 2027.
\n\n\n\nGeneral multi-agent layers differ by supervising multiple independent operational functions at once. Industrial manufacturers Kohler and Belden built this foundation through Databricks.
\n\n\n\nKohler deployed a supervisor agent coordinating demand, inventory, and planning spaces. Belden engineered a multi-tier supplier graph with task agents responding to transport incidents, targeting autonomous execution and master-data correction in subsequent phases.
\n\n\n\nSupervised autonomy requires rigid boundaries to protect capital and vendor relationships. Unchecked agents can compound errors across integrated purchase and shipping systems.
\n\n\n\nHard financial and operational tripwires should be installed before enabling direct system writes, including:
\n\n\n\nAutomated warehouse execution remains largely confined to simulation models rather than unassisted floor operations. Research by the Massachusetts Institute of Technology (MIT) and Symbotic demonstrated a 25 percent throughput increase using multi-robot path coordination inside simulated e-commerce distribution facilities.
\n\n\n\nNVIDIA released its Multi-Agent Intelligent Warehouse reference architecture to demonstrate cross-fleet planning methods. Production facilities still separate robotic movement from autonomous transaction clearing.
\n\n\n\nRohto’s expanded live-chain trial, scheduled through March 2027, is the next public test of those bounded execution loops.
\n\n\n\nLearn more about physical AI during the Physical AI Expo held in Amsterdam, London, and North America.
\n\n\n\nSee also: Gartner outlines four AI tiers in warehouse automation
\n\n\n\n![\"Banner](\"https://www.artificialintelligence-news.com/wp-content/uploads/2025/08/ai-expo-banner-2025.png\")
Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is part of TechEx and is co-located with other leading technology events including the Cyber Security & Cloud Expo. Click here for more information.
\n\n\n\nAI News is powered by TechForge Media. Explore other upcoming enterprise technology events and webinars here.
\nThe post Multi-agent AI systems are taking over supply chain execution appeared first on AI News.
","content_text":"Multi-agent AI systems are taking over supply chain execution as enterprise networks face diminishing returns from static dashboards, pushing logistics directors towards autonomous execution.\nPredictive demand models display recommendations, yet human planners still clear every action. Multi-agent systems replace that approval stage across targeted operational boundaries.\nInstead of waiting for weekly scheduling runs, independent software models ingest real-time telemetry from carrier ETAs, yard cameras, and warehouse management system events. The agents execute freight re-routing, safety stock rebalancing, and dock allocations directly inside enterprise resource software.\nLenovo\nreported this operational transition on its global iChain infrastructure across 180 markets, more than 30 factories, and 100 logistics centres. The hardware manufacturer linked an Order Fulfilment Agent and a Risk Management Agent directly to existing transaction platforms.\nAccording to Lenovo, fulfilment decisions ran three times faster, disruption response four times faster, risk assessment operated at 85 percent accuracy, and delivery accuracy increased 30 percent.\nNamed multi-agent systems on live supply chains\nA mid-size automotive parts manufacturer, documented by\nSimor Consulting\n, deployed five specialised agents across 15 countries and 200 suppliers over an 18-month production run. The company recorded an on-time delivery rise from 82 percent to 94 percent.\nThe manufacturer observed that its disruption agent detected supply threats 48 hours ahead of manual monitoring teams. Communication agents interacted smoothly with longstanding suppliers. Dialogue failed with unfamiliar vendors until the software catalogued their specific reply behaviours.\nInter-enterprise logistics routing trials show comparable results. An initial virtual-network exercise conducted by\nFujitsu\nand\nRohto Pharmaceutical\nyielded transport cost reductions of up to 30 percent. Both companies scheduled an expanded trial on Rohto’s live physical chain between January 2026 and March 2027.\nGeneral multi-agent layers differ by supervising multiple independent operational functions at once. Industrial manufacturers Kohler and Belden built this foundation through\nDatabricks\n.\nKohler\ndeployed a supervisor agent coordinating demand, inventory, and planning spaces.\nBelden\nengineered a multi-tier supplier graph with task agents responding to transport incidents, targeting autonomous execution and master-data correction in subsequent phases.\nOperational guardrails govern multi-tier agent actions\nSupervised autonomy requires rigid boundaries to protect capital and vendor relationships. Unchecked agents can compound errors across integrated purchase and shipping systems.\nHard financial and operational tripwires should be installed before enabling direct system writes, including:\nTransport rerouting scripts hold authority only within strict cost ceilings and service level agreement deltas.\nInventory adjustments that exceed predefined financial values or volume percentages automatically pause for manual planner authorisation.\nSupplier-facing communication agents remain restricted to draft modes on unvetted supplier accounts until interaction accuracy surpasses established benchmarks.\nAutomated warehouse execution remains largely confined to simulation models rather than unassisted floor operations. Research by the Massachusetts Institute of Technology (\nMIT\n) and\nSymbotic\ndemonstrated a 25 percent throughput increase using multi-robot path coordination inside simulated e-commerce distribution facilities.\nNVIDIA\nreleased its Multi-Agent Intelligent Warehouse reference architecture to demonstrate cross-fleet planning methods. Production facilities still separate robotic movement from autonomous transaction clearing.\nRohto’s expanded live-chain trial, scheduled through March 2027, is the next public test of those bounded execution loops.\nLearn more about physical AI during the\nPhysical AI Expo\nheld in Amsterdam, London, and North America.\nSee also:\nGartner outlines four AI tiers in warehouse automation\nWant to learn more about AI and big data from industry leaders?\nCheck out\nAI & Big Data Expo\ntaking place in Amsterdam, California, and London. The comprehensive event is part of\nTechEx\nand is co-located with other leading technology events including the\nCyber Security & Cloud Expo\n. Click\nhere\nfor more information.\nAI News is powered by\nTechForge Media\n. Explore other upcoming enterprise technology events and webinars\nhere\n.\nThe post\nMulti-agent AI systems are taking over supply chain execution\nappeared first on\nAI News\n.","image":{"url":"https://virtues.guru/tech-assets/99_Bqs_logo-transparent-PNG.png","file":"99_Bqs_logo-transparent-PNG.png","title":"File:Bqs logo-transparent-PNG.png"},"language":"en","translations":{"pt":{"title":"Sistemas de IA baseados em múltiplos agentes estão assumindo o controle da execução da cadeia de suprimentos.","content_text":"Sistemas de IA com múltiplos agentes estão assumindo o controle da execução da cadeia de suprimentos, substituindo painéis estáticos e planejadores humanos. Eles processam dados em tempo real e executam tarefas como a re-rotulagem de cargas, o reequilíbrio do estoque de segurança e a alocação de docas. Empresas como a Lenovo e um fabricante de peças automotivas de médio porte têm observado melhorias significativas na precisão das entregas e nas taxas de entrega no prazo. No entanto, são necessários mecanismos de segurança operacionais para evitar erros e proteger os relacionamentos com clientes e fornecedores.","summary_en":"Multi-agent AI systems are taking over supply chain execution, replacing static dashboards and human planners. They ingest real-time data and execute tasks such as freight re-routing, safety stock rebalancing, and dock allocations. Companies like Lenovo and a mid-size automotive parts manufacturer have seen significant improvements in delivery accuracy and on-time delivery rates. However, operational guardrails are necessary to prevent errors and protect capital and vendor relationships.","translated_at":"2026-09-22T06:31:04.482723+00:00","model":"translategemma:4b","summary_model":"llama3.1:8b","source":"json_summary"}}},{"source":"artificialintelligence-news","title":"How to Improve Visibility Across Your Enterprise AI Ecosystem","url":"https://www.artificialintelligence-news.com/news/how-to-improve-visibility-across-your-enterprise-ai-ecosystem/","slug":"how-to-improve-visibility-across-your-enterprise-ai-ecosystem","published_at":"2026-09-21T07:50:48+00:00","author":"WebFX","categories":["Sponsored Content"],"summary_html":"AI adoption has outpaced AI governance across enterprise environments, creating a fundamental security problem. Organisations cannot protect what they cannot see, and visibility has become the prerequisite for all other AI security controls. Traditional monitoring tools fail to track AI activity effectively, creating significant risks that require new strategies for security teams to regain control […]
\nThe post How to Improve Visibility Across Your Enterprise AI Ecosystem appeared first on AI News.
","content_html":"AI adoption has outpaced AI governance across enterprise environments, creating a fundamental security problem. Organisations cannot protect what they cannot see, and visibility has become the prerequisite for all other AI security controls. Traditional monitoring tools fail to track AI activity effectively, creating significant risks that require new strategies for security teams to regain control of their AI ecosystems.
\n\n\n\nThe gap between enterprise AI adoption and AI governance is becoming harder for security leaders to ignore. Cisco’s 2025 Cybersecurity Readiness Index found that 60% of organisations do not know the specific requests employees make to GenAI tools. That lack of visibility makes it harder to monitor data movement, enforce policy and understand which tools or agents are operating across the enterprise.
\n\n\n\nThe issue is structural, not cultural. Organisations built their monitoring tools to track traditional software, and these systems were never designed to detect how AI moves through a network in the first place. Standard discovery tools can identify a software subscription but often miss AI usage patterns entirely.
\n\n\n\nWhen employees circumvent official channels to use AI tools, IT loses visibility into where sensitive company data is actually going. This structural gap poses real operational risk, as data flows to destinations that the security team cannot monitor or control.
\n\n\n\nShadow AI refers to employees using AI tools and applications without explicit approval from the organisation. This differs from traditional shadow IT because rogue software subscriptions remain visible to standard discovery tools in ways that AI usage often does not. Each dimension of shadow AI carries a distinct risk profile and requires a different kind of response.
\n\n\n\nA common version of this risk occurs when an employee pastes a document or dataset into a public chatbot to save time on a routine task. This behavior is rarely malicious. It reflects normal workers reaching for the most convenient available tool, not an intent to bypass security protocols.
\n\n\n\nThis risk lies within the tools a company has already approved, since the original security review predates the AI features added to the platform later. Embedded capabilities are harder to catch than stand-alone tool use because the traffic looks identical to normal platform activity from a monitoring standpoint.
\n\n\n\nAgents take action within a system rather than simply answering a question, distinguishing them from most assistants. Oversight lags far behind deployment because agents are often stood up quickly to solve an immediate workflow problem without a formal review process. An agent acting with unmonitored access can affect systems and data at a speed no human review process can match, making this an urgent risk to address.
\n\n\n\nThe failure is architectural rather than a matter of insufficient effort or budget. Traditional security tools were built to track known software in expected locations, which does not match how AI capabilities actually move through an organisation.
\n\n\n\nA tool built to catalog applications has no reliable way to classify or control the behavior of an AI agent acting inside one. The monitoring systems most enterprises rely on simply lack the framework to capture AI activity patterns, creating visibility gaps that grow wider as AI adoption accelerates.
\n\n\n\nOrganisations must adopt specific strategies to bridge the visibility gap and regain control of AI activity across the enterprise. The following approaches provide the foundation for securing AI ecosystems.
\n\n\n\nA one-time audit falls short because new AI tools and features are added continuously rather than on a predictable schedule. Security teams should apply the same discipline used for cloud workloads, in which every asset is tracked as a matter of routine instead of only after an incident.
\n\n\n\nA living inventory enables security teams to maintain a current picture against which to measure new activity, rather than reconstructing it after something goes wrong. This continuous approach ensures the organisation knows which AI capabilities are available in the environment at any given time.
\n\n\n\nEffectively securing AI requires applying AI to the problem, since human review alone cannot keep pace with the volume and speed of the task. Platforms that use advanced behavioral analysis can help security teams identify unusual AI activity without relying only on known attack signatures.
\n\n\n\nDarktrace provides an example of a platform built around this approach. The company has been pioneering AI since 2013, predating the more recent wave of AI-branded security tools. Its platform uses multi-layered AI to provide visibility across the on-premise network, cloud applications, email, OT systems and endpoints.
\n\n\n\nWhat makes the approach unique is that there is no starting point or prior assumptions about what a threat looks like. The technology learns every device, user and interaction, developing an understanding of normal behavior from what it observes. This allows it to spot and thread together subtle behavioral anomalies that indicate a threat, whereas other security solutions try to predefine what constitutes a threat based on attack patterns observed in the past.
\n\n\n\nNo system or agent should be granted access based on assumed trust rather than a verified, specific need. AI agents can act across data, tools and applications. Because of this, each agent should receive only the minimum access required for its specific task.
\n\n\n\nProperly scoped access limits the damage an undetected compromise or malfunction can cause. This principle becomes especially important in AI environments where agents operate at machine speed and can propagate issues faster than human operators can respond.
\n\n\n\nVisibility remains the foundational step to safe AI innovation across the enterprise. Organisations that implement continuous discovery and advanced threat detection platforms position themselves to effectively secure their AI ecosystems. IT leaders should prioritise platforms that provide comprehensive coverage across AI touchpoints and use behavioral analysis rather than signature-based detection to identify threats in real time.
\n\n\n\n\nThe post How to Improve Visibility Across Your Enterprise AI Ecosystem appeared first on AI News.
","content_text":"AI adoption has outpaced AI governance across enterprise environments, creating a fundamental security problem. Organisations cannot protect what they cannot see, and visibility has become the prerequisite for all other AI security controls. Traditional monitoring tools fail to track AI activity effectively, creating significant risks that require new strategies for security teams to regain control of their AI ecosystems.\nThe Growing Enterprise AI Visibility Crisis\nThe gap between enterprise AI adoption and AI governance is becoming harder for security leaders to ignore. Cisco’s 2025 Cybersecurity Readiness Index found that\n60% of organisations do not know\nthe specific requests employees make to GenAI tools. That lack of visibility makes it harder to monitor data movement, enforce policy and understand which tools or agents are operating across the enterprise.\nThe issue is structural, not cultural. Organisations built their monitoring tools to track traditional software, and these systems were never designed to detect how AI moves through a network in the first place. Standard discovery tools can identify a software subscription but often miss AI usage patterns entirely.\nWhen employees circumvent official channels to use AI tools, IT loses visibility into where sensitive company data is actually going. This structural gap poses real operational risk, as data flows to destinations that the security team cannot monitor or control.\nUnderstanding the Risks of Shadow AI\nShadow AI refers to\nemployees using AI tools and applications\nwithout explicit approval from the organisation. This differs from traditional shadow IT because rogue software subscriptions remain visible to standard discovery tools in ways that AI usage often does not. Each dimension of shadow AI carries a distinct risk profile and requires a different kind of response.\nUnsanctioned Stand-Alone Tools\nA common version of this risk occurs when an employee pastes a document or dataset into a public chatbot to save time on a routine task. This behavior is rarely malicious. It reflects normal workers reaching for the most convenient available tool, not an intent to bypass security protocols.\nEmbedded Software as a Service Capabilities\nThis risk lies within the tools a company has already approved, since the original security review predates the AI features added to the platform later. Embedded capabilities are harder to catch than stand-alone tool use because the traffic looks identical to normal platform activity from a monitoring standpoint.\nAutonomous AI Agents\nAgents take action within a system rather than simply answering a question, distinguishing them from most assistants. Oversight lags far behind deployment because agents are often stood up quickly to solve an immediate workflow problem without a formal review process. An agent acting with unmonitored access can affect systems and data at a speed no human review process can match, making this an urgent risk to address.\nWhy Traditional Security Tooling Fails\nThe failure is architectural rather than a matter of insufficient effort or budget. Traditional security tools were built to track known software in expected locations, which does not match how AI capabilities actually move through an organisation.\nA tool built to catalog applications has no reliable way to classify or control the behavior of an AI agent acting inside one. The monitoring systems most enterprises rely on simply lack the framework to capture AI activity patterns, creating visibility gaps that grow wider as AI adoption accelerates.\nStrategies to Secure the AI Environment\nOrganisations must adopt specific strategies to bridge the visibility gap and regain control of AI activity across the enterprise. The following approaches provide the foundation for securing AI ecosystems.\nEstablish Continuous Discovery and Inventory\nA one-time audit falls short because new AI tools and features are added continuously rather than on a predictable schedule. Security teams should apply the same discipline used for cloud workloads, in which every asset is tracked as a matter of routine instead of only after an incident.\nA living inventory enables security teams to maintain a current picture against which to measure new activity, rather than reconstructing it after something goes wrong. This continuous approach ensures the organisation knows which AI capabilities are available in the environment at any given time.\nImplement Multi-Layered AI Threat Detection\nEffectively securing AI requires applying AI to the problem, since human review alone cannot keep pace with the volume and speed of the task. Platforms that use advanced behavioral analysis can help security teams identify unusual AI activity without relying only on known attack signatures.\nDarktrace\nprovides an example of a platform built around this approach. The company has been pioneering AI since 2013, predating the more recent wave of AI-branded security tools. Its platform uses multi-layered AI to provide visibility across the on-premise network, cloud applications, email, OT systems and endpoints.\nWhat makes the approach unique is that there is no starting point or prior assumptions about what a threat looks like. The technology learns every device, user and interaction, developing an understanding of normal behavior from what it observes. This allows it to spot and thread together subtle behavioral anomalies that indicate a threat, whereas other security solutions try to predefine what constitutes a threat based on attack patterns observed in the past.\nEnforce Zero Trust Access Controls\nNo system or agent should be granted access based on assumed trust rather than a verified, specific need. AI agents\ncan act across data\n, tools and applications. Because of this, each agent should receive only the minimum access required for its specific task.\nProperly scoped access limits the damage an undetected compromise or malfunction can cause. This principle becomes especially important in AI environments where agents operate at machine speed and can propagate issues faster than human operators can respond.\nTaking Control of the AI Future\nVisibility remains the foundational step to safe AI innovation across the enterprise. Organisations that implement continuous discovery and advanced threat detection platforms position themselves to effectively secure their AI ecosystems. IT leaders should prioritise platforms that provide comprehensive coverage across AI touchpoints and use behavioral analysis rather than signature-based detection to identify threats in real time.\nThe post\nHow to Improve Visibility Across Your Enterprise AI Ecosystem\nappeared first on\nAI News\n.","image":{"url":"https://virtues.guru/tech-assets/96_Information_Technology_Engineering.png","file":"96_Information_Technology_Engineering.png","title":"File:Information Technology Engineering.png"},"language":"en","translations":{"pt":{"title":"Como melhorar a visibilidade em todo o seu ecossistema de IA empresarial","content_text":"A adoção de IA tem superado a governança de IA, criando um problema de segurança. As organizações carecem de visibilidade sobre as atividades de IA, o que dificulta o monitoramento do fluxo de dados e a aplicação de políticas. As ferramentas tradicionais de monitoramento não conseguem rastrear as atividades de IA de forma eficaz, criando riscos significativos. Para proteger os ecossistemas de IA, as organizações devem adotar estratégias como a descoberta e inventário contínuos, a detecção de ameaças de IA em várias camadas e controles de acesso baseados em \"zero confiança\".","summary_en":"AI adoption has outpaced AI governance, creating a security problem. Organisations lack visibility into AI activity, making it hard to monitor data movement and enforce policy. Traditional monitoring tools fail to track AI activity effectively, creating significant risks. To secure AI ecosystems, organisations must adopt strategies such as continuous discovery and inventory, multi-layered AI threat detection, and zero trust access controls.","translated_at":"2026-09-22T06:23:54.022442+00:00","model":"translategemma:4b","summary_model":"llama3.1:8b","source":"json_summary"}}}]}