Latest news with #AIassistant
CNET
3 days ago
- Business
- CNET
OpenAI Unleashes ChatGPT Agent to Be Your Personal Assistant
OpenAI has launched ChatGPT Agent for Pro, Plus and Team users -- an AI-powered personal assistant that connects to your various services online to help complete tasks custom to you, the company said in a livestream on Thursday. Using the power of OpenAI's "reasoning" o3 model, which has deep research capabilities, along with Operator, ChatGPT Agent can go step-by-step from one task to the other to put together complex reports. ChatGPT Agent can visualize what it's doing via a virtual computer interface. Here, you can see its train of thought, what it's browsing on the net and what systems it's connecting to. During the livestream, OpenAI members demoed what it'd be like to use ChatGPT Agent to figure out travel plans for a wedding. Here, ChatGPT Agent was able to check the wedding site provided, find the date, find nearby hotels, what the weather might be like, browse other websites, look for tuxedos on Nordstrom's highly visual webpage, look for wedding gifts and present to you a final report. Users can also interject to ask for help finding a pair of men's black shoes. ChatGPT Agent can also connect to various services, like your email, calendar, news apps and others. OpenAI says safety is foundational to this model and that ChatGPT Agent will request your permission before doing anything sensitive, like making purchases or submitting forms. ChatGPT agent won't do "high-risk tasks" like financial transactions or legal advice. The system will also reject "harmful or illegal requests." OpenAI didn't immediately respond to a request for comment. The debut of ChatGPT Agent is the next step in the chase for artificial general intelligence, or AGI. These are AI models so advanced that they would have a general understanding of the world around them and could outcompete humans in tasks. AI systems now, while seeming smart, actually have no understanding of the world and simply arrange words mathematically in a manner that best mimics human speech based on certain criteria. Agentic AI systems, ones that can go out and do complex tasks across various systems, have been seen as a major step towards achieving AGI. Because agentic AI can better mimic complex human train of thought, these systems can be given a task and can be sent off to get them accomplished. While it's hard to measure the economic impact of billions of AI agents doing the work once done by humans, entire industries will be affected.
Associated Press
4 days ago
- Business
- Associated Press
VerifiID Launches AI Compliance Copilot to Streamline Hiring and I-9 Verification
VerifiID unveils a new platform and AI Copilot to help HR teams automate I-9, E-Verify, and onboarding compliance—faster, safer, and at scale. 'The Compliance Copilot isn't just AI—it's a trusted assistant that helps HR teams move faster, stay compliant, and onboard with confidence.'— Mo Saeed, Founder and CEO of BOSTON, MA, UNITED STATES, July 16, 2025 / / -- VerifiID Launches AI-Powered Compliance Copilot and New Platform to Simplify Hiring at Scale VerifiID, a fast-growing leader in workforce compliance and onboarding automation, today announced the official launch of its redesigned platform, along with its newest innovation: the Compliance Copilot — a multilingual, AI-powered assistant that helps HR teams manage hiring compliance with speed and clarity. Built for employers, staffing firms, and high-turnover industries, delivers a faster, smarter, and more secure way to manage background checks, I-9s, E-Verify, and employee documentation — all from one place. The Compliance Copilot is VerifiID's standout feature: an always-on assistant trained to answer complex compliance questions in real-time, across 50+ languages. From 'How do I complete an I-9 remotely?' to 'Do I need E-Verify in Florida?', Copilot empowers HR teams to stay compliant without delays or legal guesswork. 'This launch is a major milestone in our mission to make hiring safer, faster, and more accessible,' said Mo Saeed, Founder and CEO of VerifiID. 'The Compliance Copilot isn't just a chatbot. It's a trusted assistant for HR — one that understands the regulations and helps teams move with confidence.' Key platform features include: I-9 and E-Verify workflows ID and background checks with real-time verification document upload and storage 4.24/7 Copilot support for compliance guidance APIs and enterprise scalability Unlike traditional tools, VerifiID combines automation with human-centered AI to remove friction from the hiring process — particularly for industries like hospitality, healthcare, logistics, and retail, where speed and scale are critical. 'This was a team effort from day one,' said Zubeir, CTO of VerifiID. 'We didn't just build a tool. We built an experience where compliance becomes intuitive and scalable.' Employers can use VerifiID to invite new hires, complete identity verification remotely (no video calls required), and auto-generate fully compliant Form I-9s — including remote hiring flows using biometric-based verification. Every step is tracked, stored, and audit-ready. The platform also supports multilingual onboarding and document compliance, helping HR teams onboard diverse, distributed, and seasonal workforces with ease. VerifiID's Compliance Copilot is available now — free to use for HR teams looking for quick answers and faster onboarding support. The platform is live at Media Contact: Pablo Alverado Head of Customer Success [email protected] Mo Saeed +1 617-970-1467 email us here Visit us on social media: LinkedIn Legal Disclaimer: EIN Presswire provides this news content 'as is' without warranty of any kind. We do not accept any responsibility or liability for the accuracy, content, images, videos, licenses, completeness, legality, or reliability of the information contained in this article. If you have any complaints or copyright issues related to this article, kindly contact the author above.
Geeky Gadgets
4 days ago
- Business
- Geeky Gadgets
Google Memory Bank Released : Long-Term AI Memory for Your Agents
What if your AI assistant could truly remember you—your preferences, your habits, even the context of past conversations? With the release of Google's new Memory Bank, this vision is no longer a distant dream but a tangible reality. Designed to equip AI agents with long-term memory capabilities, Memory Bank addresses one of the most persistent challenges in artificial intelligence: the inability to recall and build upon past interactions. This innovation promises to transform AI systems from reactive tools into proactive, context-aware companions, capable of delivering personalized and seamless experiences. Imagine an AI that doesn't just respond but evolves with you—this is the future Google is unveiling. In this update, MG explores how Memory Bank redefines what's possible for AI by introducing adaptive memory storage and retrieval. You'll discover why traditional stateless models fall short in personalization and continuity, and how Google's solution bridges that gap with features like prospective reflection and retrospective refinement. Whether you're a developer eager to integrate this technology or simply curious about its potential, Memory Bank's capabilities open up a world of possibilities for smarter, more intuitive AI systems. As we delve deeper, consider this: what could it mean for technology to truly remember you? Google's Memory Bank Unveiled Why Memory Bank Matters Memory Bank is a response to the challenges posed by traditional AI memory systems. Stateless models, while effective for single-session tasks, are inherently limited in their ability to maintain continuity across multiple interactions. This lack of continuity often results in repetitive or impersonal responses. Memory Bank bridges this gap by allowing AI agents to store and retrieve relevant information over extended periods. This capability significantly enhances personalization, allowing AI systems to adapt to individual user preferences and deliver contextually appropriate responses. For example, an AI-powered virtual assistant equipped with Memory Bank can remember a user's preferences for specific services or products, making sure a more tailored and engaging experience. Limitations of Existing AI Memory Systems Traditional approaches to AI memory management often fail to deliver the efficiency and relevance required for modern applications. Common methods, such as storing entire conversation histories or using similarity searches to retrieve past interactions, come with significant drawbacks: Inefficient: Storing large volumes of data increases operational costs and slows down processing times, making these systems impractical for large-scale applications. Storing large volumes of data increases operational costs and slows down processing times, making these systems impractical for large-scale applications. Error-Prone: Retrieval mechanisms frequently surface irrelevant or outdated information, leading to inconsistent user experiences. Retrieval mechanisms frequently surface irrelevant or outdated information, leading to inconsistent user experiences. Rigid: These systems lack the flexibility to adapt to evolving user behavior, limiting their ability to refine memory retrieval processes effectively. Memory Bank addresses these shortcomings by introducing a more intelligent and scalable approach to memory management. Its design ensures that only the most relevant and valuable information is stored and retrieved, optimizing both performance and user satisfaction. Long-Term Memory for Your AI Agent Watch this video on YouTube. Here are more detailed guides and articles that you may find helpful on long-term memory for AI. Key Features of Memory Bank Memory Bank incorporates several innovative features that set it apart from traditional memory systems. These features are designed to enhance the functionality and adaptability of AI agents: Personalization: Automatically identifies and stores user-specific preferences and interactions, allowing AI systems to deliver responses that are tailored to individual needs. Automatically identifies and stores user-specific preferences and interactions, allowing AI systems to deliver responses that are tailored to individual needs. Continuity: Selectively stores and retrieves only the most relevant portions of past interactions, making sure that conversations remain seamless and contextually coherent across sessions. Selectively stores and retrieves only the most relevant portions of past interactions, making sure that conversations remain seamless and contextually coherent across sessions. Adaptability: Employs reinforcement learning techniques to refine memory retrieval processes over time, improving the relevance and accuracy of responses. Employs reinforcement learning techniques to refine memory retrieval processes over time, improving the relevance and accuracy of responses. Cohesive Memory Creation: Consolidates fragmented session data into unified, meaningful memories, enhancing the AI's ability to understand and anticipate user needs. These features collectively enable AI agents to deliver more intelligent, responsive, and user-centric interactions, making Memory Bank a valuable tool for developers and businesses alike. How Memory Bank Works Memory Bank is seamlessly integrated with the Google Cloud Platform (GCP) and the Google Agent Development Kit (ADK), making it accessible to developers across a wide range of industries. Its functionality is supported by REST API integration, making sure compatibility with various frameworks and systems. Developers using Google ADK gain access to several key benefits: Automatic memory storage and retrieval capabilities, reducing the need for manual intervention. Seamless integration into existing AI workflows, minimizing disruption to ongoing projects. Minimal coding effort required for implementation, allowing for faster deployment. This flexibility ensures that Memory Bank can be easily adopted by developers, allowing them to enhance their AI systems with long-term memory capabilities without significant overhead. Technical Innovations Memory Bank introduces two new techniques that redefine how AI systems manage and use memory: Prospective Reflection: This technique consolidates fragmented session data into cohesive memory summaries, allowing AI agents to maintain a clear and organized understanding of past interactions. By creating structured memories, the system ensures that relevant information is readily accessible when needed. This technique consolidates fragmented session data into cohesive memory summaries, allowing AI agents to maintain a clear and organized understanding of past interactions. By creating structured memories, the system ensures that relevant information is readily accessible when needed. Retrospective Reflection: By analyzing user interactions over time, this method refines the relevance of retrieved information, making sure that future responses are more accurate and contextually appropriate. This adaptive learning process enhances the overall user experience. These innovations allow Memory Bank to go beyond simple data storage, allowing AI systems to evolve and improve in response to user behavior and preferences. Real-World Applications The introduction of Memory Bank opens up a wide range of practical applications for AI agents across various industries. Some notable use cases include: Personalized Experiences: AI systems equipped with Memory Bank can recall user-specific details, such as preferences, past queries, or frequently used services, to deliver a more tailored and engaging experience. AI systems equipped with Memory Bank can recall user-specific details, such as preferences, past queries, or frequently used services, to deliver a more tailored and engaging experience. Improved Continuity: By maintaining context across multiple interactions, Memory Bank ensures that users receive consistent and relevant responses, which is particularly valuable in fields like customer support, education, and healthcare. By maintaining context across multiple interactions, Memory Bank ensures that users receive consistent and relevant responses, which is particularly valuable in fields like customer support, education, and healthcare. Enhanced Decision-Making: In industries such as finance or logistics, Memory Bank can help AI systems analyze historical data to provide more informed recommendations and predictions. These capabilities make Memory Bank a powerful tool for businesses seeking to use AI for consistent, personalized, and context-aware interactions. Getting Started with Memory Bank Implementing Memory Bank is a straightforward process for developers working within the Google ecosystem. To get started, you need to create an agent engine session within GCP. The system can be accessed via REST API or natively integrated with Google ADK. Key benefits of this setup include: Automated storage and retrieval of long-term memory, reducing the need for manual configuration. Scalability to handle large datasets and complex interactions, making it suitable for enterprise-level applications. Ease of integration into existing AI workflows, allowing rapid deployment and minimal disruption. This streamlined approach ensures that developers can quickly enhance their AI systems with advanced memory capabilities, unlocking new possibilities for innovation and user engagement. Performance and Efficiency Extensive benchmark testing has demonstrated that Memory Bank outperforms traditional long-term memory systems in both efficiency and relevance. Its ability to adapt to user behavior and refine memory retrieval processes ensures a superior user experience. These advancements make Memory Bank an essential tool for developers aiming to create context-aware AI agents that deliver consistent and meaningful interactions. By addressing the limitations of stateless AI models and introducing a scalable, intelligent approach to memory management, Memory Bank represents a significant advancement in the field of artificial intelligence. Media Credit: MG Filed Under: AI, Top News Latest Geeky Gadgets Deals Disclosure: Some of our articles include affiliate links. If you buy something through one of these links, Geeky Gadgets may earn an affiliate commission. Learn about our Disclosure Policy.
Yahoo
5 days ago
- Business
- Yahoo
How one cybersecurity startup is using AI to keep clients safe
Artificial intelligence has become a key part of cybersecurity. It's helping cybersecurity companies keep their customers' data safe, but, unfortunately, it's also a new tool for criminals to exploit. Arctic Wolf ( CEO and president Nick Schneider joins Asking for a Trend with Josh Lipton to discuss how his company, valued at more than $4 billion, is leveraging AI to fight those who want to use the technology for harm. To watch more expert insights and analysis on the latest market action, check out more Asking for a Trend here. Let's talk about AI, a trend investors are very excited about. How is your company leveraging that tech? Yeah, so there's two primary ways. First of all, AI's been used in cybersecurity for a long time. There's two primary ways you can use it kind of in the back end or in the core of the platform to be able to make sense of all the data that you're ingesting. So, for example, our business is ingesting now somewhere in the neighborhood of 9 trillion security events a week. But our average customer is only really actioning, gonna call it two to three of those alerts per week. And really the magic between those two numbers is all through automation and AI, in AI. Now, that's not really customer facing. That's a lot of what's happening in the back end. The second way to leverage it is through LLMs or engagement with the end user of themselves. So, like, for example, we have a product called Cypher. It's an AI assistant. And effectively what that allows end customers to do is understand the data that they're getting through the platform to be able to take action within their environment or do something within their business. And we talked about how you're using AI. How are the bad guys using it? Yeah, the easiest one to like really wrap your head around is through like fishing campaigns, right? So it used to be a world where a fishing campaign was actually pretty easy to, you know, point out if you were an individual and you were paying attention. Now they've gotten pretty sophisticated, right? So the emails are written really well. They use logos that are really well depicted, the headers that they leverage for the emails that they send are really hard to decipher, like a one versus an I, for example. So so they've gotten really sophisticated with leveraging AI and then, you know, also leveraging the AI in a bulk fashion to do that to, you know, thousands of people at once as opposed to trying to kind of do it in a piecemeal fashion. When you talk about bad guys, Nick, who are we talking about in broad strokes? Yeah. Yeah, I got this question today, too. So um, so bad actors in general, generally speaking they're in groups of bad actors and I think that's where we see the majority of the activity. So those can be actually organizations in certain countries, or individuals now with the advent of AI can do this from their home. Uh but then you also see in certain industries concerned over like nation states, right? So if you're in critical infrastructure, anything that might be, you know, critical to a country or an organization that would be interesting to a nation state, you'll see some nation-state activity as well.
GSM Arena
12-07-2025
- GSM Arena
vivo X200 FE review
GSMArena Team , 11 July 2025 . The vivo X200 FE boots Android 15 with a layer of the company's in-house Funtouch OS on top, also v.15. The X200 FE will be eligible for 4 major OS release updates plus an extra year with security updates, which sounds like a good level of futureproofing. The latest Funtouch isn't much different from previous iterations in look and feel, and almost gives off a bit of a dated vibe with its visuals. Funtouch 15 The present-day functionality is, of course, all there. Funtouch includes niceties now taken for granted like large folders, and the split-screen/pop-up window implementation is about as straightforward as it gets, too. The app drawer is worth mentioning too, because that's where Funtouch keeps its widget selection for some reason. Funtouch 15 The X200 FE also comes with Google's Circle to search on board and the Gemini AI assistant as well. A Google Lens-powered screen translation feature is also present. Google AI You also get live translation of phone calls, an AI helper in the Notes app, and an AI Transcript Assist utility for making summaries out of spoken conversations. Note that all of these may be region-dependent and not present in your locale. AI Transcript Assist • AI in Note Also, there's some AI-powered functionality in the gallery editor. You get to delete objects, remove people, and fight reflections - all of those with varying success, of course. AI in Albums Like other vivos, the X200 FE comes with this unique feature where it can copy and emulate a physical access card. It can scan and copy your RFID card as long as it's based on NFC technology (13.56MHz). It doesn't work for 125KHz cards. NFC Performance and benchmarks The vivo X200 FE is powered by the flagship-grade Dimensity 9300+ 4nm chip. It packs an 8-core processor with a 3.25 GHz Cortex-X4 prime core, three 2.85 GHz Cortex-X4 cores, and four 2.0GHz Cortex-A720 cores. These are all prime and high-performance cores, there are no A510 power-efficient ones. The GPU is the Immortalis-G720 MC12. There are several memory configurations - a base 12GB/256GB version, our review unit's 12GB/512GB spec, as well as the 16GB/512 top-spec variant. Not all of these will be available in all markets, as usual. The vivo X200 FE posted excellent benchmark scores, bested only by the most recent top-notch Snapdragon models. The X100 Pro from last year, which had a similar Dimensity inside, posted slightly higher scores altogether, with the regular X200 and its newer Dimensity 9400 being yet a little more potent, but the FE is plenty powerful enough. The phone kept 53% of its maximum CPU performance when running our usual stress test, and it dropped to this level after just 5 mins. The GPU sustained performance is better - the phone kept about 60% of its graphics capabilities during the 3D Mark stress test. CPU test • GPU test Overall, the vivo X200 FE enters the flagship class with its Dimensity 9300+ chip, scoring below the cream of the crop, but better than the premium midrangers out there. Its sustained performance is average at best, with the phone becoming quite hot after 10-15 mins running at peak performance. Still, for a compact device with powerful hardware, we'd say it's managed rather well. Android 15 with Funtouch 15 The vivo X200 FE boots Android 15 with a layer of the company's in-house Funtouch OS on top, also v.15. The X200 FE will be eligible for 4 major OS release updates plus an extra year with security updates, which sounds like a good level of futureproofing. The latest Funtouch isn't much different from previous iterations in look and feel, and almost gives off a bit of a dated vibe with its visuals. Funtouch 15 The present-day functionality is, of course, all there. Funtouch includes niceties now taken for granted like large folders, and the split-screen/pop-up window implementation is about as straightforward as it gets, too. The app drawer is worth mentioning too, because that's where Funtouch keeps its widget selection for some reason. Funtouch 15 The X200 FE also comes with Google's Circle to search on board and the Gemini AI assistant as well. A Google Lens-powered screen translation feature is also present. Google AI You also get live translation of phone calls, an AI helper in the Notes app, and an AI Transcript Assist utility for making summaries out of spoken conversations. Note that all of these may be region-dependent and not present in your locale. AI Transcript Assist • AI in Note Also, there's some AI-powered functionality in the gallery editor. You get to delete objects, remove people, and fight reflections - all of those with varying success, of course. AI in Albums There is also this fan-favorite feature - emulation of physical access cards. This way you phone would also become your keychain, or maybe wallet. NFC Performance and benchmarks The vivo X200 FE is powered by the flagship-grade Dimensity 9300+ 4nm chip. It packs an 8-core processor with a 3.25 GHz Cortex-X4 prime core, three 2.85 GHz Cortex-X4 cores, and four 2.0GHz Cortex-A720 cores. These are all prime and high-performance cores, there are no A510 power-efficient ones. The GPU is the Immortalis-G720 MC12. There are several memory configurations - a base 12GB/256GB version, our review unit's 12GB/512GB spec, as well as the 16GB/512 top-spec variant. Not all of these will be available in all markets, as usual. The vivo X200 FE posted very good benchmark scores, bested only by the most recent top-notch Snapdragon models. The X100 Pro from last year that has a similar Dimensity inside posted slightly higher scores altogether, with the regular X200 and its newer Dimensity 9400 being yet a little more potent, but the FE is plenty powerful enough. The phone kept 53% of its maximum CPU performance when running our usual stress test, and it dropped to this level after just 5 mins. The GPU sustained performance is better - the phone kept about 60% of its graphics capabilities during the 3D Mark stress test. CPU test • GPU test Overall, the vivo X200 FE enters the flagship class with its Dimensity 9300+ chip, scoring below the cream of the crop, but better than the premium midrangers out there. Its sustained performance is average at best, with the phone becoming quite hot after 10-15 mins running at peak performance. Still, for a compact device with powerful hardware, we'd say it's managed rather well.
