# On-device AI Published articles for On-device AI. This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles. ## CrowdStrike Accelerates Real-Time Data Classification with On-Device AI DevFeed: [CrowdStrike Accelerates Real-Time Data Classification with On-Device AI]() Original publisher: [Read original article]() Author: Lior Ribak Published: 2026-09-17T01:38:53.400452Z Content type: article Language: en Sources: [Blog]() Topics: [On-device AI](), [sensitive data](), [Language models](), [Endpoint security](), [Hardware](), [intel]() Tags: [data-security](), [endpoint-security](), [intel](), [language-models](), [npu](), [on-device-ai]() ### AI overview CrowdStrike and Intel introduced a Falcon Data Security capability that uses language models running on-device on dedicated AI hardware to classify sensitive data. The article explains that local inference is intended to avoid cloud latency and keep sensitive customer data on the endpoint while meeting real-time protection requirements. ### Source excerpt CrowdStrike worked closely with Intel to introduce a new capability in Falcon Data Security that classifies sensitive data using language models that run on-device using dedicated hardware for AI. Learn more! ## On-Device AI Series (Part 5): LiteRT-LM DevFeed: [On-Device AI Series (Part 5): LiteRT-LM]() Original publisher: [Read original article]() Author: Oğuzhan Aslan Published: 2026-09-14T05:59:12Z Content type: tutorial Language: en Sources: [ProAndroidDev - Medium]() Topics: [LiteRT](), [On-device AI](), [Large Language Model](), [Inference](), [Streaming](), [GPU](), [LLM evaluation / benchmarking](), [PyTorch]() Tags: [ai](), [android](), [android-development](), [benchmarking](), [gpu](), [inference](), [litert](), [llama](), [llm](), [multimodal](), [on-device](), [on-device-ai](), [programming](), [pytorch](), [qwen](), [software-development](), [streaming]() ### AI overview This tutorial explains LiteRT-LM for running large language models on-device. It covers the Engine/Session API, streaming output, system prompts, tool calling, multimodal inputs, thinking mode, and CPU-versus-GPU benchmarking. The article also discusses tradeoffs involving privacy, network independence, latency, memory, sampling configuration, and model capability compared with cloud APIs. ### Source excerpt Put your phone in airplane mode. Open the app, type a question, and watch the answer arrive one token at a time -- no spinner waiting on a network round-trip, no API key, no per-token bill, and nothing you typed ever leaving the device. LiteRT-LM removes the genuinely hard parts of running an LLM on-device -- KV-cache management, token streaming, backend selection -- but it doesn't remove your job so much as relocate it. What's left on your plate is a short, specific list: sizing a combined input+output token budget, owning your own sampling defaults, hand-building system prompts and tool calling out of raw text, and one native-library collision that presents as a SIGSEGV rather than a build error. Know those going in and the API itself is a clean three-step pattern. We'll get there in that order: Why you'd choose this runtime and what it costs you versus the cloud. The Engine/Session model you need to read the code at all. Real implementation samples -- streaming, system prompts and tool calling, multimodal inputs, thinking mode, and CPU-vs-GPU benchmarking. The anti-patterns to avoid. A developer-friendliness rating on the same rubric as Parts 1-4. Why Use LiteRT-LM? You reach for LiteRT-LM instead of hand-rolling generation on top of raw LiteRT when: You need multi-turn conversation, not single-shot inference -- session state and KV-cache bookkeeping are handled for you, and resetting a conversation is a session swap, not a model reload. You need streaming output -- token-by-token delivery for a responsive chat UI, instead of a blocking call that returns everything at once. You're choosing between CPU and GPU per device -- the explicit backend parameter turns that into a runtime decision instead of a build-time guess. You want a pre-converted model without doing your own PyTorch-to-LiteRT conversion work -- the Model Zoo covers Gemma, Qwen, Llama, and more out of the box. You're willing to own sampling -- the engine won't pick sane decoding defaults for you; that's on the ## Dell Pro 14 Premium Review: A 2.55-Pound Magnesium Flagship With Tandem OLED DevFeed: [Dell Pro 14 Premium Review: A 2.55-Pound Magnesium Flagship With Tandem OLED]() Original publisher: [Read original article]() Author: Dylan Bryan Published: 2026-09-02T00:35:43Z Content type: article Language: en Sources: [StorageReview.com]() Topics: [dell](), [Security](), [cpu](), [intel](), [GPU](), [On-device AI]() Tags: [consumer](), [dell](), [display](), [enterprise](), [gpu](), [intel](), [memory](), [mobile](), [on-device-ai](), [review](), [security](), [storage](), [workstation]() ### AI overview This review examines the Dell Pro 14 Premium, a lightweight 14-inch commercial notebook designed for highly mobile professionals. It emphasizes portability, recycled magnesium construction, enterprise security and manageability, replaceable battery serviceability, Tandem OLED display quality, and on-device AI capabilities. Its integrated graphics and processor options are suited to office productivity and collaboration rather than GPU rendering or CAD workloads. ### Source excerpt The Dell Pro 14 Premium is what Dell calls its ultimate commercial notebook: a 14-inch, 2.54-pound magnesium flagship aimed at highly mobile professionals rather than workstation users. It sits above the Dell Pro and Dell Pro Plus lines, and the pitch is craftsmanship and portability with enterprise security and manageability underneath, not raw compute. The The post Dell Pro 14 Premium Review: A 2.55-Pound Magnesium Flagship With Tandem OLED appeared first on StorageReview.com. ## Running golf swing analysis on an Android device with a fine-tuned Gemma 4 model DevFeed: [Running golf swing analysis on an Android device with a fine-tuned Gemma 4 model]() Original publisher: [Read original article]() Published: 2026-08-29T23:00:00Z Content type: tutorial Language: en Sources: [John O'Reilly]() Topics: [Kotlin Multiplatform](), [gemma4](), [LiteRT](), [On-device AI](), [Fine-tuning]() Tags: [ai](), [android](), [cross-platform](), [fine-tuning](), [gemma](), [gemma-4](), [kotlin](), [kotlin-multiplatform](), [litert](), [model](), [multiplatform](), [on-device-ai]() ### AI overview This article explains how FormAI adds Android on-device golf-swing analysis using a small Gemma 4 model fine-tuned to imitate Gemini for a narrow coaching task. It covers generating training data with Gemini, fine-tuning with LoRA, converting the model for LiteRT-LM, and the current fallback to cloud analysis on other platforms. ### Source excerpt FormAI is a Kotlin Multiplatform app that analyses a video of your golf swing, basketball shot or running form and gives you coaching feedback. Up to now that has always meant uploading the video to Gemini and getting the response back over the network. We've added an option to do the golf swing analysis entirely on an Android device instead, using a small Gemma 4 model that we fine-tuned to imitate Gemini for that one task. LiteRT-LM, the runtime we use for this, is itself cross platform (Android, iOS, desktop and web), but we've only wired up the Android side so far, so this path lives in androidMain and the other targets report it as unavailable and fall back to the cloud. ## Codename One Adds On-Device AI and Loopback MCP Support DevFeed: [Codename One Adds On-Device AI and Loopback MCP Support]() Original publisher: [Read original article]() Author: Shai Almog Published: 2026-08-02T00:00:00Z Content type: release Language: en Sources: [CodeName One]() Topics: [On-device AI](), [Model Context Protocol](), [Artificial Intelligence](), [LiteRT](), [Inference](), [Mobile](), [supply-chain-security]() Tags: [applications](), [inference](), [litert](), [mcp](), [mobile](), [on-device](), [on-device-ai](), [tensorflow-lite]() ### AI overview Codename One has added on-device vision, language, and LiteRT inference to its core, alongside a guarded loopback MCP transport for inspecting and operating applications. The article explains platform support, asynchronous OCR, model sessions, secure model downloads, and privacy limitations of local inference. ### Source excerpt Codename One now exposes on-device vision, language, and LiteRT inference in the core, while a guarded loopback MCP transport lets an LLM inspect and drive real applications. ## Accelerating Gemini Nano models on Pixel with frozen Multi-Token Prediction DevFeed: [Accelerating Gemini Nano models on Pixel with frozen Multi-Token Prediction]() Original publisher: [Read original article]() Published: 2026-06-26T18:30:00Z Content type: article Language: en Sources: [The latest research from Google]() Topics: [AI Inference](), [On-device AI](), [Google](), [LLMs](), [gemma]() Tags: [ai](), [efficiency](), [energy](), [gemini](), [inference](), [machine-intelligence](), [mobile-systems](), [natural-language-processing](), [on-device](), [on-device-ai](), [phones]() ### AI overview Google Research describes a method for retrofitting Multi-Token Prediction onto frozen Gemini Nano v3 production models to accelerate on-device inference on Pixel phones. The approach targets mobile energy and memory constraints, improving the speed and energy efficiency of features such as notification summaries and proofreading without requiring separate drafting models. ### Source excerpt Machine Intelligence ## Gemini in Apple's Foundation Models framework DevFeed: [Gemini in Apple's Foundation Models framework]() Original publisher: [Read original article]() Author: Miguel Ramos; Ryan Wilson Published: 2026-06-09T09:41:00Z Content type: article Language: en Sources: [Firebase Blog]() Topics: [Firebase](), [foundation-models](), [Artificial Intelligence](), [API](), [Swift](), [Inference]() Tags: [ai](), [ai-logic](), [app-check](), [apple](), [firebase](), [foundation-models](), [gemini](), [hybrid-inference](), [ios](), [news](), [on-device-ai](), [swift](), [updates](), [wwdc]() ### AI overview Firebase describes a preview integration that lets developers use cloud-hosted Gemini models through Apple's Foundation Models framework. The shared API allows switching between on-device Apple models and Gemini, while Firebase AI Logic and App Check provide request routing and attestation for Apple apps. ### Source excerpt News, tutorials, and updates from the Firebase team. ## Why I Think On-Device AI Changes Mobile Architecture DevFeed: [Why I Think On-Device AI Changes Mobile Architecture]() Original publisher: [Read original article]() Author: Akansha Jain Published: 2026-06-03T17:56:53Z Content type: opinion Language: en Sources: [Flutter Community - Medium]() Topics: [Artificial Intelligence](), [On-device AI](), [Mobile](), [Flutter](), [Android](), [data]() Tags: [ai](), [android](), [api](), [artificial-intelligence](), [dart](), [data](), [flutter](), [llm](), [mobile](), [on-device](), [on-device-ai]() ### AI overview This opinion article argues that on-device AI changes mobile application architecture by reducing dependence on network APIs. It discusses connectivity, per-call costs, and user-data concerns, drawing on a Flutter camera-demo crash involving a model already loaded on an Android phone. ### Source excerpt What a crashed app taught me about on-device AI -- and why I think you should care. I was building the demo I'd been planning for weeks. Flutter app running. Model loaded. Chat screen looking clean. I opened the camera -- the feature I was most excited about -- pointed it at something on my desk, and the app just crashed. Not a Dart exception. Not a red error screen. The whole thing just died. I stared at my Redmi Note 7 Pro for a second, opened it again, tried once more. Same result. Every single time I launched the camera with the model already loaded, Android killed my app. It took me longer than I'd like to admit to understand why. And when I finally did, I realised I'd been thinking about AI in mobile apps in a way that was quietly wrong for a long time. Before that, let me back up. For the past couple of years, whenever I wanted to add an AI feature to an app, the answer was pretty obvious: use an API. Need a chatbot? API. Need to analyze an image? API. Need text summarisation? API. A few lines of Dart, a request goes out, and something surprisingly intelligent comes back. It worked well enough that I stopped questioning the architecture behind it. Every one of those API calls travels over the network, gets processed on hardware you don't own, and comes back. Which means three things are always true, and always have been: The feature needs internet. Which is fine until you're in a metro tunnel, on a flight, or somewhere with patchy coverage. Your app just stops being smart. Users notice -- and they don't file bug reports, they just leave. You pay for every call. At low volume it's fine. At scale, it genuinely isn't. 10,000 users x 50 queries a day x per-token pricing. At low volume you barely notice it. At scale, those costs become a product decision. User data leaves the device. The health app where someone is logging symptoms. The finance app where they're describing their expenses. The journal app where they're writing things they wouldn't say out loud. For tho ## Expanding on-device AI in Microsoft Edge: New models and APIs for the web DevFeed: [Expanding on-device AI in Microsoft Edge: New models and APIs for the web]() Original publisher: [Read original article]() Author: Patrick Brosset Published: 2026-06-02T16:28:06Z Content type: article Language: en Sources: [Microsoft Edge Blog]() Topics: [Edge](), [On-device AI](), [Artificial Intelligence](), [API](), [Language models](), [Web](), [JavaScript](), [Microsoft](), [Cloud](), [Open Source](), [cpu](), [hugging face]() Tags: [ai](), [ai-apis](), [api](), [availability](), [cloud](), [cpu](), [github](), [gpu](), [javascript](), [microsoft-edge](), [models](), [on-device-ai](), [open-source](), [web](), [web-apis](), [web-platform]() ### AI overview Microsoft Edge is expanding its on-device AI capabilities with a developer preview of the smaller Aion-1.0-Instruct language model, new Language Detector and Translator APIs in Edge 148, and experimental on-device speech recognition through the Web Speech API. The updates target broader device support, including CPU-only systems, while emphasizing privacy, network independence, and web-developer experimentation. ### Source excerpt At Build 2025, we introduced the Prompt and Writing Assistance APIs in Microsoft Edge with the Phi-4-mini language model. Since then, we' The post Expanding on-device AI in Microsoft Edge: New models and APIs for the web appeared first on Microsoft Edge Blog. ## Private analytics via zero-trust aggregation DevFeed: [Private analytics via zero-trust aggregation]() Original publisher: [Read original article]() Published: 2026-05-27T16:56:00Z Content type: article Language: en Sources: [The latest research from Google]() Topics: [Security, Privacy and Abuse Prevention](), [Security](), [Zero Trust](), [Google](), [On-device AI](), [Artificial Intelligence](), [Android]() Tags: [ai](), [android](), [cryptographic](), [on-device-ai](), [privacy](), [security](), [security-privacy-and-abuse-prevention](), [zero-trust]() ### AI overview Google Research introduces a private analytics solution that combines cryptographic secure aggregation with the transparency and attestation properties of trusted execution environments. The design follows a zero-trust principle so Google can obtain only anonymized, aggregated population insights while individual user data remains protected. The article discusses applications of federated analytics and on-device technologies, including Android SafetyCore, Pixel Recorder, and Gboard. ### Source excerpt Security, Privacy and Abuse Prevention ## What's New in Android Security and Privacy in 2026 DevFeed: [What's New in Android Security and Privacy in 2026]() Original publisher: [Read original article]() Author: Eugene Liderman Published: 2026-05-12T17:00:00Z Content type: article Language: en Sources: [Security]() Topics: [Android](), [Android Security](), [Security](), [spoofing](), [Social engineering](), [threat detection](), [On-device AI](), [Chrome]() Tags: [ai](), [android](), [android-security](), [fraud](), [none](), [on-device-ai](), [security](), [social-engineering](), [spoofing](), [threat-detection]() ### AI overview The article describes planned Android security and privacy enhancements for 2026, including verified financial calls to combat spoofed banking scams. Android can verify incoming calls through participating financial apps and automatically end calls that are not genuine. It also highlights expanded Live Threat Detection, which uses on-device AI to analyze app behavior and warn about suspicious activity. ### Source excerpt New Android security and privacy features ## Android expands AI-powered scam detection to Samsung devices and more regions DevFeed: [Android expands AI-powered scam detection to Samsung devices and more regions]() Original publisher: [Read original article]() Author: Edward Fernandez (noreply@blogger.com) Published: 2026-02-25T15:17:00Z Content type: release Language: en Sources: [Google Online Security]() Topics: [Android](), [Google AI](), [Google](), [Mobile]() Tags: [ai](), [android](), [android-security](), [devices](), [google](), [mobile](), [none](), [on-device-ai](), [protection](), [scams](), [smartphones]() ### AI overview Google describes Android's AI-powered scam protections, including Scam Detection warnings for suspicious calls. The article announces expansion of call Scam Detection to Samsung devices and additional regions, while noting existing availability on Google Pixel devices in several countries. ### Source excerpt Posted by Lyubov Farafonova, Product Manager, Phone by Google; Alberto Pastor Nieto, Sr. Product Manager Google Messages and RCS Spam and Abuse We've shared how Android's proactive, multi-layered scam defenses utilize Google AI to protect users around the world from over 10 billion suspected malicious calls and messages every month1. While that scale is significant, the true impact of these protections is best understood through the stories of the individuals they help keep safe every day. This includes people like Majik B., an IT professional in Sunnyvale, California. Despite his technical background, Majik recently found himself on a call that felt dangerously legitimate. While using his Pixel, he received a call that appeared to be from his bank. The number looked correct, the caller knew his name and his address, and the story about a "suspicious charge" made perfect sense. "I'm usually pretty careful about this stuff," Majik recalled, "but I stayed on the line longer than I normally would. Even knowing how these scams work, it was convincing in the moment." The turning point came when his phone displayed a Scam Detection warning during the call, which provided a critical moment to pause and reflect. Majik hung up, checked his bank app directly, and confirmed there was no fraudulent charge. For Majik, Scam Detection was the intervention he needed: "The warning is what made me pause and avoid a bad situation". While stories like Majik's show how our existing protections provide a robust shield against scams, our work isn't done. As scammers evolve their tactics and create more convincing and personalized threats, we're using the best of Google AI to stay one step ahead. A recent evaluation by Counterpoint Research found that Android smartphones provide the most comprehensive AI-powered protections of any mobile platform. We are committed to building on this foundation by expanding our AI-powered protections to more users and devices, while rolling out new feature ## Introducing Gemma 3n: The developer guide DevFeed: [Introducing Gemma 3n: The developer guide]() Original publisher: [Read original article]() Author: Omar Sanseviero; Ian Ballantyne Published: 2025-10-25T17:54:47Z Content type: release Language: en Sources: [Google DeepMind News]() Topics: [Local AI]() Tags: [ai](), [developer](), [gemma](), [hugging-face](), [inference](), [llama-cpp](), [on-device](), [on-device-ai](), [release](), [transformer-architecture]() ### AI overview Gemma 3n is released as a mobile-first, multimodal model for on-device AI. The guide introduces its MatFormer architecture, pre-extracted E4B and E2B models, and tooling for fine-tuning and deployment. ### Source excerpt Gemma 3n is designed for the developer community that helped shape Gemma. ## AI application topologies: cloud, edge, local, and hybrid inference DevFeed: [AI application topologies: cloud, edge, local, and hybrid inference]() Original publisher: [Read original article]() Author: Alex Ewerlöf Published: 2025-10-24T21:15:00Z Content type: article Language: en Sources: [Alex Ewerlof Notes]() Topics: [Artificial Intelligence](), [Cloud](), [Local AI](), [Inference](), [On-device AI](), [Latency](), [Software as a service]() Tags: [ai](), [cloud](), [edge](), [edge-ai](), [inference](), [latency](), [local-ai](), [on-device-ai](), [privacy](), [saas]() ### AI overview The article categorizes AI application topologies by where inference compute occurs relative to data: cloud, edge, local, and hybrid. It describes trade-offs involving capability, latency, cost, privacy, connectivity, vendor limits, and centralized-service outages. ### Source excerpt Cloud AI, Edge AI, Local AI, and Hybrid AI ## Understanding hybrid inference: the role of Firebase AI Logic DevFeed: [Understanding hybrid inference: the role of Firebase AI Logic]() Original publisher: [Read original article]() Author: Ankita Saxena Published: 2025-10-15T10:00:00Z Content type: tutorial Language: en Sources: [Firebase Blog]() Topics: [Firebase](), [Artificial Intelligence](), [Inference](), [Web app]() Tags: [ai](), [ai-logic](), [chrome](), [firebase](), [firebase-ai-logic](), [gemini](), [hybrid-inference](), [inference](), [multimodal](), [offline](), [on-device](), [on-device-ai](), [real-time](), [tutorials](), [web](), [web-app]() ### AI overview This Firebase tutorial explains hybrid inference, where AI features use on-device models when available and fall back to cloud-hosted models when needed. It demonstrates web app use cases including text summarization, offline image captioning, and real-time transcription and translation. ### Source excerpt News, tutorials, and updates from the Firebase team. ## Smart photo organization through privacy-first, on-device AI DevFeed: [Smart photo organization through privacy-first, on-device AI]() Original publisher: [Read original article]() Author: Cynthia Wang Published: 2025-10-15T08:00:00Z Content type: tutorial Language: en Sources: [Firebase Blog]() Topics: [On-device AI](), [Firebase](), [PWA](), [SDKs](), [web applications](), [browser]() Tags: [ai](), [ai-logic](), [firebase](), [gemini](), [hybrid-inference](), [on-device](), [on-device-ai](), [privacy](), [pwa](), [sdks](), [tutorials](), [web](), [web-browser]() ### AI overview This Firebase tutorial explains how to build a privacy-focused photo organization progressive web app using on-device AI. It describes running a local model in the browser so photos remain on the user's device and can be analyzed offline, with Firebase AI Logic SDKs detecting and caching compatible models such as Gemini Nano and providing a fallback approach for other browsers. ### Source excerpt News, tutorials, and updates from the Firebase team. ## Introducing: Local Browser AI DevFeed: [Introducing: Local Browser AI]() Original publisher: [Read original article]() Author: Alex Ewerlöf Published: 2025-10-13T07:51:29Z Content type: article Language: en Sources: [Alex Ewerlof Notes]() Topics: [browser](), [Artificial Intelligence](), [Browser Extension](), [JavaScript](), [On-device AI]() Tags: [ai](), [browser](), [extension](), [interface](), [javascript](), [on-device-ai](), [open-source]() ### AI overview This technical post explains how the Prompt API enables local language-model chat in the browser and examines the Local Browser AI extension, which is built with plain JavaScript and native browser APIs. It also describes the extension's permissions, privacy properties, open-source MIT license, and supported platforms. ### Source excerpt Using the new Prompt API for local chat in the browser ## Adding AI-powered star-rating suggestions to reviews with Firebase AI Logic DevFeed: [Adding AI-powered star-rating suggestions to reviews with Firebase AI Logic]() Original publisher: [Read original article]() Author: Alexander Nohe; Jeff Huleatt Published: 2025-10-01T00:00:00Z Content type: tutorial Language: en Sources: [Firebase Blog]() Topics: [Firebase](), [Artificial Intelligence](), [Google AI](), [API keys](), [Generative AI]() Tags: [ai](), [ai-logic](), [api-keys](), [app-check](), [firebase](), [firestore](), [gemini](), [hybrid-inference](), [mcp](), [on-device-ai](), [react](), [recaptcha](), [security-rules](), [tutorials]() ### AI overview This tutorial explains how the Friendly Eats app uses Firebase AI Logic to suggest a star rating while a user writes a review. It discusses choosing a client-side SDK, protecting Gemini API keys through a proxy, using on-device models when available, securing requests with Firebase App Check, and balancing model cost, speed, and quality. ### Source excerpt News, tutorials, and updates from the Firebase team. ## New Firebase AI Logic features to explore - September 2025 updates DevFeed: [New Firebase AI Logic features to explore - September 2025 updates]() Original publisher: [Read original article]() Author: Miguel Ramos; Rachel Saunders Published: 2025-09-10T00:00:00Z Content type: release Language: en Sources: [Firebase Blog]() Topics: [Firebase](), [Artificial Intelligence](), [SDKs](), [On-device AI](), [Image](), [API](), [Web](), [Android](), [Flutter](), [Unity](), [Chrome]() Tags: [ai](), [ai-logic](), [android](), [api](), [app-check](), [debug](), [firebase](), [firebase-ai-logic](), [flutter](), [gemini](), [hybrid-inference](), [image](), [images](), [inference](), [js](), [news](), [observability](), [offline](), [on-device](), [on-device-ai](), [unity](), [updates]() ### AI overview Firebase AI Logic's September 2025 updates add Gemini Live API voice interactions, client-side image editing and generation with Imagen, Nano Banana image workflows, hybrid on-device inference for web apps, and additional monitoring and observability capabilities. ### Source excerpt News, tutorials, and updates from the Firebase team. ## Android pKVM Achieves SESIP Level 5 Security Certification DevFeed: [Android pKVM Achieves SESIP Level 5 Security Certification]() Original publisher: [Read original article]() Author: Edward Fernandez (noreply@blogger.com) Published: 2025-08-12T16:00:00Z Content type: release Language: en Sources: [Google Online Security]() Topics: [Android](), [Android Security](), [Security](), [open-source-security](), [Google](), [virtualization](), [On-device AI]() Tags: [android](), [android-security](), [google](), [integrity](), [none](), [on-device-ai](), [open-source-security](), [security](), [standard](), [virtualization](), [vulnerability]() ### AI overview Google announces that protected KVM (pKVM), the hypervisor powering the Android Virtualization Framework, achieved SESIP Level 5 certification after evaluation by Dekra against the TrustCB SESIP scheme. The article says this certification is intended to support highly critical isolated workloads, including on-device AI handling personalized data, and provide a common open-source security foundation for Android device manufacturers. ### Source excerpt Posted by Dave Kleidermacher, VP Engineering, Android Security & Privacy Today marks a watershed moment and new benchmark for open-source security and the future of consumer electronics. Google is proud to announce that protected KVM (pKVM), the hypervisor that powers the Android Virtualization Framework, has officially achieved SESIP Level 5 certification. This makes pKVM the first software security system designed for large-scale deployment in consumer electronics to meet this assurance bar. Supporting Next-Gen Android Features The implications for the future of secure mobile technology are profound. With this level of security assurance, Android is now positioned to securely support the next generation of high-criticality isolated workloads. This includes vital features, such as on-device AI workloads that can operate on ultra-personalized data, with the highest assurances of privacy and integrity. This certification required a hands-on evaluation by Dekra, a globally recognized cybersecurity certification lab, which conducted an evaluation against the TrustCB SESIP scheme, compliant to EN-17927. Achieving Security Evaluation Standard for IoT Platforms (SESIP) Level 5 is a landmark because it incorporates AVA_VAN.5, the highest level of vulnerability analysis and penetration testing under the ISO 15408 (Common Criteria) standard. A system certified to this level has been evaluated to be resistant to highly skilled, knowledgeable, well-motivated, and well-funded attackers who may have insider knowledge and access. This certification is the cornerstone of the next-generation of Android's multi-layered security strategy. Many of the TEEs (Trusted Execution Environments) used in the industry have not been formally certified or have only achieved lower levels of security assurance. This inconsistency creates a challenge for developers looking to build highly critical applications that require a robust and verifiable level of security. The certified pKVM changes this p ## Introducing hybrid on-device inference using Firebase AI Logic DevFeed: [Introducing hybrid on-device inference using Firebase AI Logic]() Original publisher: [Read original article]() Author: Siddharth Gupta; Erik Eldridge Published: 2025-06-09T00:00:00Z Content type: release Language: en Sources: [Firebase Blog]() Topics: [Firebase](), [On-device AI](), [Inference](), [AI Models](), [Web]() Tags: [ai](), [ai-logic](), [experimental](), [feature](), [firebase](), [gemini](), [hybrid-inference](), [inference](), [javascript](), [model](), [on-device](), [on-device-ai](), [web]() ### AI overview Firebase introduces an experimental hybrid inference feature in its AI Logic client SDK for Web. The feature uses available on-device models such as Gemini Nano in Chrome and falls back to cloud-hosted Gemini models when needed, supporting online and offline AI features. ### Source excerpt News, tutorials, and updates from the Firebase team. ## Real-Time AI Sound Generation on Arm: A Personal Tool for Creative Freedom DevFeed: [Real-Time AI Sound Generation on Arm: A Personal Tool for Creative Freedom]() Original publisher: [Read original article]() Author: EricSondhi; Michael G Published: 2025-06-03T15:04:36Z Content type: article Language: en Sources: [Hugging Face - Blog]() Topics: [Arm](), [On-device AI](), [Generative AI](), [cpu](), [PyTorch](), [AI Inference](), [Open Source](), [hugging face]() Tags: [ai](), [arm](), [audio](), [content](), [generative-ai](), [hugging-face](), [music](), [on-device-ai](), [privacy](), [pytorch]() ### AI overview A software engineer and music producer describes a personal sound-generation app that runs entirely on-device on Arm-based CPUs. It uses the Stable Audio Open model, PyTorch, and TorchAudio to generate sounds from text prompts within seconds, then places the output directly into an Ableton Live project workflow. The approach emphasizes CPU efficiency, low latency, privacy, and creative ownership without requiring cloud inference. ### Source excerpt As a software engineer and music producer, I'm always exploring how technology can expand creative expression. That curiosity recently led me to build a personal sound generation app that runs directly on-device--powered by an Arm-based CPU and open-source generative AI models. It's fast, private, and enables me to generate studio-ready sounds from a simple prompt, all within seconds. ## Building AI-powered apps with Firebase AI Logic DevFeed: [Building AI-powered apps with Firebase AI Logic]() Original publisher: [Read original article]() Author: Thevi Sundaralingam; Dmitry Lyalin; Miguel Ramos Published: 2025-05-20T10:49:00Z Content type: release Language: en Sources: [Firebase Blog]() Topics: [Firebase](), [Generative AI](), [SDKs](), [Genkit](), [API]() Tags: [ai](), [ai-logic](), [dashboards](), [firebase](), [firebase-ai-logic](), [gemini](), [genkit](), [google-i-o](), [hybrid-inference](), [inference](), [launch](), [node-js](), [on-device](), [on-device-ai](), [sdks](), [vertex-ai]() ### AI overview Firebase announces Firebase AI Logic, the evolution of Vertex AI in Firebase, with client-side and Genkit-based server-side integrations for generative AI. New capabilities include Gemini Developer API access, hybrid and on-device inference, Unity support, image generation and editing, enhanced observability, and dynamic model lookup in Genkit for Node.js. ### Source excerpt News, tutorials, and updates from the Firebase team. ## On-device AI may shape the future of Android phone development DevFeed: [On-device AI may shape the future of Android phone development]() Original publisher: [Read original article]() Author: Kaushik Gopal Published: 2025-02-09T18:53:19Z Content type: opinion Language: en Sources: [Kaushik Gopal's Site]() Topics: [Artificial Intelligence](), [On-device AI](), [Mobile](), [Android](), [Google]() Tags: [ai](), [android](), [development](), [gemini](), [google](), [mobile](), [models](), [on-device](), [on-device-ai](), [phones](), [samsung]() ### AI overview This commentary predicts that on-device AI, including smaller models running locally on phones, will become a major focus of Android development. It also suggests Google or Samsung could gain mobile market share as Apple struggles to establish a strong AI position. ### Source excerpt In Vinay's latest newsletter, he asks a few of us #AndroidDev to predict what the future of Android development is going to look like. Yours truly had this as one of the predictions: AI everything 🙄 ... On the product side, we'll see more on-device AI, with smaller models like Gemini Flash/o3-mini running locally to provide operator-like intelligence directly on phones and this will probably be what most folks are geared towards doing for mobile development. Looking at this Youtube video that's now doing the rounds, I'm starting to feel ever so slightly validated. Google (or Samsung) truly have their shot of gaining significant Mobile market share again, especially given how much Apple has been floundering in the AI space. [Next page]()