Key takeaways: AI and automation are transforming e-commerce by delivering personalized experiences and streamlining operations like inventory management. Mobile-first strategies and social media platforms are becoming key sales channels, driving purchases and engagement. Ethical practices, sustainability, and enhanced customer support are shaping brand loyalty, as consumers increasingly seek transparency and responsible business practices. E-commerce isn’t just about selling products online anymore — it’s about keeping up with your customers’ ever-evolving expectations. With technology advancing rapidly and shopping habits shifting constantly, businesses that adapt to these changes will thrive, while those that don’t risk falling behind. As an online business owner, staying on top of the latest trends in e-commerce is your competitive edge. I’ve identified 10 key e-commerce trends you need to know to keep your business ahead of the curve in 2025: AI and automation 1. AI-driven personalization Artificial intelligence is transforming e-commerce by enabling hyper-personalized shopping experiences. AI analyzes customer data — like browsing habits and purchase behavior — to deliver tailored product recommendations and marketing. A 2024 ACA study1 reports that 75% of customers prefer businesses offering personalized experiences, making this technology crucial for staying competitive. 💡Tip: Use AI tools to personalize product recommendations and email campaigns based on customer browsing and purchase history. ⚙️Tech tools to explore: Platforms Optimizely or Dynamic Yield both offer robust AI solutions that allow you to personalize product recommendations, on-site content, and even email marketing based on customer behaviors, such as browsing and purchasing history. 2. Inventory and fulfillment automation Automation is transforming how businesses handle inventory and fulfillment, making processes faster and more accurate. Imagine using AI tools to predict demand trends — this helps you stock just the right amount, avoiding the headaches of overstocking or running out of popular items. Companies like Amazon are taking it a step further with robots in their warehouses, sorting, packing, and shipping orders in record time. As online shopping grows, automated solutions will be essential for processing higher order volumes efficiently. By 2027, the global warehouse automation market is expected to grow from $23.44 billion to $41 billion2, making automation a key driver of e-commerce scalability and customer satisfaction. 💡Tip: Invest in inventory management software that integrates automation tools for demand forecasting and real-time tracking. This ensures optimal stock levels, reduces manual errors, and supports scalability during peak shopping seasons. For fulfillment, explore automated solutions, such as robotic picking systems or partnering with advanced fulfillment centers, to enhance delivery speed and accuracy. ⚙️Tech tools to explore: Peak Inventory AI provides AI-powered advanced inventory forecasting, while Shopify e-commerce sites can benefit from Monocle AI for data-driven sales & inventory forecasts. Finally Robotic is a robotic grocery fulfillment solution that can help increase productivity and fulfillment accuracy. 3. Price optimization Dynamic pricing, fueled by AI, is becoming a game-changer for e-commerce. AI tools analyze market trends, competitor pricing, and customer behavior in real time to set optimal prices for maximum profitability. For example, Amazon uses dynamic pricing to adjust millions of prices daily, ensuring competitiveness and driving sales. This strategy helps businesses stay agile in fluctuating markets while maximizing revenue. 💡Tip: To leverage price optimization effectively, start by adopting AI-powered pricing tools that monitor competitor prices, market demand, and customer behavior. Additionally, make sure to analyze your own data to identify patterns — such as price sensitivity based on customer segments — and adjust accordingly. For example, if you notice that customers are more likely to purchase during off-peak times, your AI tool could automatically lower prices during those hours, increasing conversions without sacrificing profit margins. ⚙️Tech tools to explore: Peak Inventory AI provides AI-powered advanced inventory forecasting, while Shopify e-commerce sites can benefit from Monocle AI for data-driven sales & inventory forecasts. Finally Robotic Tools like Prisync or Dynamic Pricing AI can provide real-time insights into when and how to adjust your prices, ensuring you remain competitive and maximize revenue. 4. Personalized pricing Tailoring prices based on customer segments or behavior is gaining traction. Personalized pricing uses data such as browsing history or loyalty status to offer exclusive discounts or dynamic offers. For instance, returning customers might see lower prices on repeat purchases, which increases the likelihood of conversion. This strategy not only boosts customer loyalty but also enhances the shopping experience, making pricing feel more customer-centric. 💡Tip: To Use customer data like purchase history, browsing behavior, or loyalty program status to offer tailored discounts or personalized promotions. ⚙️Tech tools to explore: Reactev is a dynamic pricing platform that uses AI to adjust pricing based on real-time market demand and consumer behavior. For Shopify merchants, the Pricing.AI app personalizes prices for different customer segments. Mobile commerce 5. Mobile shopping takes the lead As the use of smartphones continues to increase, mobile commerce is expected to rise. By 2025, mobile commerce is expected to have a 59% share of total retail e-commerce sales worldwide3. Optimized mobile apps and responsive websites help businesses tap into this growing audience, ensuring a seamless shopping experience. 💡Tip: To Ensure your website is mobile-optimized by using responsive design and fast-loading pages. Consider developing a user-friendly mobile app with features like one-click checkout and personalized recommendations to make the mobile shopping experience seamless and engaging. ⚙️Tech tools to explore: Consider tools like Shopify’s Mobile App Builder to create seamless, customized mobile shopping experiences, or Omnisend or a comprehensive marketing automation solution that includes SMS, email, and push notifications. Google’s Mobile-Friendly Test can help you ensure your website is optimized for mobile users, while Appy Pie can assist in developing an app with features like one-click checkout. 6. Mobile wallet adoption Digital wallets, such as Apple Pay and Google Pay, are transforming payments with the convenience and security they offer to consumers. Digital wallet transactions are expected to grow 73%, from $10 trillion in 2024 to $17 trillion in 20294, driven by their widespread adoption for online and in-store purchases. This surge reflects shifting consumer preferences for seamless and contactless payment methods that prioritize speed and data protection. Businesses that

As software-defined wide-area networks (SD-WANs) hit the peak of excitement, Forrester analysts noticed that the hype wasn’t turning into deployments. Many of the SD-WAN investigations were being held up by security teams that put a pause on the projects as the teams tried to wrap their heads around the shift in security architecture and controls from data center to cloud. At the time, networking and security were becoming interdependent, yet organizations and solutions were not quite there. Forrester put together a model that would help clients build a single, integrated networking and security model, coining a term that we thought was fitting — Zero Trust edge (ZTE) — as these changes made way for providing Zero Trust security via networking at the edge. But we weren’t the only ones naming this same new phenomenon; the term secure access service edge (SASE) rose around the same time. It didn’t set the bar quite at the same threshold as ZTE, but it described the same change. At the time, the market was full of partnerships between networking and security vendors, and only a few vendors had started to create a single solution with management and monitoring delivered from the cloud. Why Now? Five years later, the market is full of integrated solutions. It is also clear at this time that SASE (pronounced “sassy”) is the preferred market term. As such, we are going to switch over to this market-preferred term while pushing that term to meet the same bar we set for ZTE. For me, this is an easy decision, as customer-first has always been my one key truth to live by. Why? Please indulge me as I take a trip down memory lane, as my career is shaped by two prior experiences: Engineering aircraft components. Straight out of college, I spent time at a small aircraft company engineering new capabilities and improving the manufacturing, safety, or quality of existing parts. In that world, a slight change in dimension of component, such as landing-gear tube thickness, could cause a crash. I spent a lot of time talking to the pilots to see what improvements would drive them to buy the next model or lure new buyers. My greatest contribution: a curved dashboard. The costs of manufacturing the dashboard increased, but current and potential clients loved it. It made the aircraft feel personalized, and the instruments were easier to read. Launching ProVision ASIC and 5400. While I was at HP ProCurve (before the HP split into HPE and HP Inc.), account managers heavily pushed client visits to promote the previous launch (5400 and ProVision ASIC). Not only did I get a lot of direct feedback from customers about the new products (and launch) that helped shape the next cycle, but they also weren’t shy about sharing their thoughts about other products. Many of these thoughts were about real barriers they faced that made their day-to-day or real-world scenarios challenging. Hence, I try to create and release information as if I’m walking in the customer’s shoes, removing any hurdles to getting the best information to make the right decision for their organizations. If clients search “SASE” looking for best practices, design guides, and vendor comparisons, then that is the term we must use to get them what they need. Rather than waste cycles drawing comparisons between the two, we will simply push SASE to be better and achieve a higher standard to meet clients’ needs. We will be pivoting our original definition to now define SASE this way: A solution that combines security and networking functionalities — such as software-defined wide-area networks (SD-WANs), cloud access security brokers (CASBs), Zero Trust network access (ZTNA), and secure web gateways (SWGs) — delivered and supported by a single vendor with any combination of cloud, software, or hardware components. Tactically, this means that over the next few weeks, our current research with ZTE will be relabeled with SASE, along with any future research listed. A big thank you to my friend and former colleague, David Holmes, for his collaboration on this research. It was truly a pleasure. source

By Daniel Wilson ( February 19, 2025, 7:55 PM EST) — Former U.S. Department of Defense official Katie Arrington, a key figure in establishing its Cybersecurity Maturity Model Certification program who previously left the Pentagon after a contentious suspension, announced she has rejoined the DOD as chief information security officer…. Law360 is on it, so you are, too. A Law360 subscription puts you at the center of fast-moving legal issues, trends and developments so you can act with speed and confidence. Over 200 articles are published daily across more than 60 topics, industries, practice areas and jurisdictions. A Law360 subscription includes features such as Daily newsletters Expert analysis Mobile app Advanced search Judge information Real-time alerts 450K+ searchable archived articles And more! Experience Law360 today with a free 7-day trial. source

Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn More When I was a kid there were four AI agents in my life. Their names were Inky, Blinky, Pinky and Clyde and they tried their best to hunt me down. This was the 1980s and the agents were the four colorful ghosts in the iconic arcade game Pac-Man. By today’s standards they weren’t particularly smart, yet they seemed to pursue me with cunning and intent. This was decades before neural networks were used in video games, so their behaviors were controlled by simple algorithms called heuristics that dictate how they would chase me around the maze.   Most people don’t realize this, but the four ghosts were designed with different “personalities.” Good players can observe their actions and learn to predict their behaviors. For example, the red ghost (Blinky) was programmed with a “pursuer” personality that charges directly towards you. The pink ghost (Pinky) on the other hand, was given an “ambusher” personality that predicts where you’re going and tries to get there first. As a result, if you rush directly at Pinky, you can use her personality against her, causing her to actually turn away from you. I reminisce because in 1980 a skilled human could observe these AI agents, decode their unique personalities and use those insights to outsmart them. Now, 45 years later, the tides are about to turn. Like it or not, AI agents will soon be deployed that are tasked with decoding your personality so they can use those insights to optimally influence you. The future of AI manipulation In other words, we are all about to become unwitting players in “The game of humans” and it will be the AI agents trying to earn the high score. I mean this literally — most AI systems are designed to maximize a “reward function” that earns points for achieving objectives. This allows AI systems to quickly find optimal solutions. Unfortunately, without regulatory protections, we humans will likely become the objective that AI agents are tasked with optimizing.  I am most concerned about the conversational agents that will engage us in friendly dialog throughout our daily lives. They will speak to us through photorealistic avatars on our PCs and phones and soon, through AI-powered glasses that will guide us through our days. Unless there are clear restrictions, these agents will be designed to conversationally probe us for information so they can characterize our temperaments, tendencies, personalities and desires, and use those traits to maximize their persuasive impact when working to sell us products, pitch us services or convince us to believe misinformation. This is called the “AI Manipulation Problem,” and I’ve been warning regulators about the risk since 2016. Thus far, policymakers have not taken decisive action, viewing the threat as too far in the future. But now, with the release of Deepseek-R1, the final barrier to widespread deployment of AI agents — the cost of real-time processing — has been greatly reduced. Before this year is out, AI agents will become a new form of targeted media that is so interactive and adaptive, it can optimize its ability to influence our thoughts, guide our feelings and drive our behaviors. Superhuman AI ‘salespeople’ Of course, human salespeople are interactive and adaptive too. They engage us in friendly dialog to size us up, quickly finding the buttons they can press to sway us. AI agents will make them look like amateurs, able to draw information out of us with such finesse, it would intimidate a seasoned therapist. And they will use these insights to adjust their conversational tactics in real-time, working to persuade us more effectively than any used car salesman. These will be asymmetric encounters in which the artificial agent has the upper hand (virtually speaking). After all, when you engage a human who is trying to influence you, you can usually sense their motives and honesty. It will not be a fair fight with AI agents. They will be able to size you up with superhuman skill, but you won’t be able to size them up at all. That’s because they will look, sound and act so human, we will unconsciously trust them when they smile with empathy and understanding, forgetting that their facial affect is just a simulated façade.  In addition, their voice, vocabulary, speaking style, age, gender, race and facial features are likely to be customized for each of us personally to maximize our receptiveness. And, unlike human salespeople who need to size up each customer from scratch, these virtual entities could have access to stored data about our backgrounds and interests. They could then use this personal data to quickly earn your trust, asking you about your kids, your job or maybe your beloved New York Yankees, easing you into subconsciously letting down your guard. When AI achieves cognitive supremacy To educate policymakers on the risk of AI-powered manipulation, I helped in the making of an award-winning short film entitled Privacy Lost that was produced by the Responsible Metaverse Alliance, Minderoo and the XR Guild. The quick 3-minute narrative depicts a young family eating in a restaurant while wearing autmented reality (AR) glasses. Instead of human servers, avatars take each diner’s orders, using the power of AI to upsell them in personalized ways. The film was considered sci-fi when released in 2023 — yet only two years later, big tech is engaged in an all-out arms race to make AI-powered eyewear that could easily be used in these ways. In addition, we need to consider the psychological impact that will occur when we humans start to believe that the AI agents giving us advice are smarter than us on nearly every front. When AI achieves a perceived state of “cognitive supremacy” with respect to the average person, it will likely cause us to blindly accept its guidance rather than using our own critical thinking. This deference to a perceived superior intelligence (whether truly superior

By Christopher Cole ( February 20, 2025, 6:50 PM EST) — A pro-consumer group urged the Eleventh Circuit to revisit a ruling last month that overturned the Federal Communications Commission’s requirement that individual businesses obtain each consumer’s consent to contact them through comparison shopping sites…. Law360 is on it, so you are, too. A Law360 subscription puts you at the center of fast-moving legal issues, trends and developments so you can act with speed and confidence. Over 200 articles are published daily across more than 60 topics, industries, practice areas and jurisdictions. A Law360 subscription includes features such as Daily newsletters Expert analysis Mobile app Advanced search Judge information Real-time alerts 450K+ searchable archived articles And more! Experience Law360 today with a free 7-day trial. source

Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn More Large language models (LLMs) may have changed software development, but enterprises will need to think twice about entirely replacing human software engineers with LLMs, despite OpenAI CEO Sam Altman’s claim that models can replace “low-level” engineers. In a new paper, OpenAI researchers detail how they developed an LLM benchmark called SWE-Lancer to test how much foundation models can earn from real-life freelance software engineering tasks. The test found that, while the models can solve bugs, they can’t see why the bug exists and continue to make more mistakes.  The researchers tasked three LLMs — OpenAI’s GPT-4o and o1 and Anthropic’s Claude-3.5 Sonnet — with 1,488 freelance software engineer tasks from the freelance platform Upwork amounting to $1 million in payouts. They divided the tasks into two categories: individual contributor tasks (resolving bugs or implementing features), and management tasks (where the model roleplays as a manager who will choose the best proposal to resolve issues).  “Results indicate that the real-world freelance work in our benchmark remains challenging for frontier language models,” the researchers write.  The test shows that foundation models cannot fully replace human engineers. While they can help solve bugs, they’re not quite at the level where they can start earning freelancing cash by themselves.  Benchmarking freelancing models The researchers and 100 other professional software engineers identified potential tasks on Upwork and, without changing any words, fed these to a Docker container to create the SWE-Lancer dataset. The container does not have internet access and cannot access GitHub “to avoid the possible of models scraping code diffs or pull request details,” they explained. The team identified 764 individual contributor tasks, totaling about $414,775, ranging from 15-minute bug fixes to weeklong feature requests. These tasks, which included reviewing freelancer proposals and job postings, would pay out $585,225. The tasks were added to the expensing platform Expensify.  The researchers generated prompts based on the task title and description and a snapshot of the codebase. If there were additional proposals to resolve the issue, “we also generated a management task using the issue description and list of proposals,” they explained. From here, the researchers moved to end-to-end test development. They wrote Playwright tests for each task that applies these generated patches which were then “triple-verified” by professional software engineers. “Tests simulate real-world user flows, such as logging into the application, performing complex actions (making financial transactions) and verifying that the model’s solution works as expected,” the paper explains.  Test results After running the test, the researchers found that none of the models earned the full $1 million value of the tasks. Claude 3.5 Sonnet, the best-performing model, earned only $208,050 and resolved 26.2% of the individual contributor issues. However, the researchers point out, “the majority of its solutions are incorrect, and higher reliability is needed for trustworthy deployment.” The models performed well across most individual contributor tasks, with Claude 3.5-Sonnet performing best, followed by o1 and GPT-4o.  “Agents excel at localizing, but fail to root cause, resulting in partial or flawed solutions,” the report explains. “Agents pinpoint the source of an issue remarkably quickly, using keyword searches across the whole repository to quickly locate the relevant file and functions — often far faster than a human would. However, they often exhibit a limited understanding of how the issue spans multiple components or files, and fail to address the root cause, leading to solutions that are incorrect or insufficiently comprehensive. We rarely find cases where the agent aims to reproduce the issue or fails due to not finding the right file or location to edit.” Interestingly, the models all performed better on manager tasks that required reasoning to evaluate technical understanding. These benchmark tests showed that AI models can solve some “low-level” coding problems and can’t replace “low-level” software engineers yet. The models still took time, often made mistakes, and couldn’t chase a bug around to find the root cause of coding problems. Many “low-level” engineers work better, but the researchers said this may not be the case for very long.  source

With their outdated technology and high costs, legacy codebases hold enterprises back. But in many cases, the prospect of migrating to modern cloud native, open source languages1  seems even worse. Migration projects typically require significant time and money. Artificial intelligence (AI) tools have emerged to help, but many businesses fear they will expose their intellectual property, hallucinate errors or fail on large codebases because of their prompt limits. And that leaves them with daunting manual migrations on their hands. EXL Code Harbor is a GenAI-powered, multi-agent tool that enables the fast, accurate migration of legacy codebases while addressing these crucial concerns. How Code Harbor works Code Harbor accelerates current state assessment, code transformation and optimization, and code testing and validation. Throughout each stage of the process, it relies on task-specific, finely tuned agents built to exceed the efficiency and expertise of humans.                                     Assessment: Deciphers and documents the business logic, dependencies and functionality of legacy code. Governance: Maps data flows, dependencies, and transformations across different systems. Migration: Chunks, simplifies, and moves the code to the desired language, then recompiles it to recreate its full logic. Optimizes code. Testing & Validation: Auto-generates test data when real data is unavailable, ensuring robust testing environments. Execute both legacy and converted code in parallel to identify runtime issues. Auto-corrects errors iteratively, flagging only critical issues for human review. To accomplish this, Code Harbor uses AI agents that replicate the entire migration journey. By taking EXL’s expertise in helping enterprises design both legacy and modern architectures and building it into these agents, the tool tackles every migration task with greater accuracy and efficiency: Business Analyst: Code explanation, documentation, pseudo code.     Code Assessor Agent: Workload analysis, dependency identification, and code composition and complexity. Data Navigator Agent: Maps data flows, dependencies, and transformations. Code Simplifier Agent: Logically chunks the source code. Code Migrator Agent: Translates legacy code while preserving functionality. Code Optimizer Agent: Enhances the code’s readability and refactors its logic. Synthetic Data Generator Agent: Ensures robust testing by creating test data. Code Executor Agent: Executes legacy and converted code in parallel. Debugger Agent: Autocorrects and flags errors. Data Validator Agent: Compares the source and migrated code outputs. Code Harbor’s orchestration layer automates workflows across these agents, allowing enterprises to lessen their risk by migrating code in phases. Real-world benefits  By eliminating redundant code and optimizing performance, Code Harbor reduces the manual effort required for assessment, conversion and testing by 60% to 80%. Let’s take a deeper look at the tool’s benefits: Increased performance and efficiency: Code Harbor analyzes the business context of code and utilizes the benefits of the target language instead of performing a line-by-line migration. Every agent is designed to enhance code efficiency, maintainability and adaptability at every stage of the process. Greater integration and scalability: This modular architecture distributes tasks across multiple agents working in parallel, so Code Harbor can perform more work in less time. And by chunking and simplifying code pre-migration, the tool ensures enterprises can move and manage even the largest codebases. Improved security and compliance: Code Harbor doesn’t store or use customer data for training or fine-tuning models, and businesses can choose which large language models to use to best protect their data: open source or proprietary, on-premises or in the cloud. Because the tool generates synthetic testing data gleaned from the context of the code, it’s able to optimize, debug and test code in a way that other tools simply can’t — all while protecting the privacy of an enterprise’s actual data and IP. It further avoids IP infringement by training AI models that are trained on coding data with permissive licenses. Backed by 20 years of expertise and a purpose-built multi-agent framework, Code Harbor provides the flexible, customer-centric approach that businesses need for complex code migrations. Banks see faster migrations Enterprises in the financial services industry are already reaping the benefits. A top UK bank worked with EXL to move its extract, transform, load codebase from SAS to Python. The migration, which included complete testing and number matching, was over 35% faster than previous manual migrations. In another case, a global financial institution used Code Harbor to optimize its BigQuery syntax, logic and explainability. EXL also generated synthetic data to use for iterative testing and other code improvements. Overall, the firm cut code compute time and memory usage by 25% and query plans by 10%. By using GenAI and a security-first approach, Code Harbor accelerates modern migrations, freeing enterprises to focus on innovation. To learn more about how it can benefit your organization, attend the upcoming webinar, AI in Action: Driving the Shift to Scalable AI. source

By Jared Foretek ( February 21, 2025, 5:43 PM EST) — A Georgia-based phone and internet provider is appealing a $429,000 recovery order from the Federal Communications Commission’s Wireline Competition Bureau over the company’s alleged failure to verify subscribers qualified for pandemic-era subsidies, arguing that it is being punished for using the eligibility verification system that the commission itself requires them to use…. Law360 is on it, so you are, too. A Law360 subscription puts you at the center of fast-moving legal issues, trends and developments so you can act with speed and confidence. Over 200 articles are published daily across more than 60 topics, industries, practice areas and jurisdictions. A Law360 subscription includes features such as Daily newsletters Expert analysis Mobile app Advanced search Judge information Real-time alerts 450K+ searchable archived articles And more! Experience Law360 today with a free 7-day trial. source

Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn More When DeepSeek-R1 first emerged, the prevailing fear that shook the industry was that advanced reasoning could be achieved with less infrastructure. As it turns out, that’s not necessarily the case. At least, according to Together AI, the rise of DeepSeek and open-source reasoning has had the exact opposite effect: Instead of reducing the need for infrastructure, it is increasing it. That increased demand has helped fuel the growth of Together AI’s platform and business. Today the company announced a $305 million series B round of funding, led by General Catalyst and co-led by Prosperity7. Together AI first emerged in 2023 with an aim to simplify enterprise use of open-source large language models (LLMs). The company expanded in 2024 with the Together enterprise platform, which enables AI deployment in virtual private cloud (VPC) and on-premises environments. In 2025, Together AI is growing its platform once again with reasoning clusters and agentic AI capabilities.  The company claims that its AI deployment platform has more than 450,000 registered developers and that the business has grown 6X overall year-over-year. The company’s customers include enterprises as well as AI startups such as  Krea AI, Captions and Pika Labs. “We are now serving models across all modalities: language and reasoning and images and audio and video,” Vipul Prakash, CEO of Together AI, told VentureBeat. The huge impact DeepSeek-R1 is having on AI infrastructure demand DeepSeek-R1 was hugely disruptive when it first debuted, for a number of reasons — one of which was the implication that a leading edge open-source reasoning model could be built and deployed with less infrastructure than a proprietary model. However, Prakash explained, Together AI has grown its infrastructure in part to help support increased demand of DeepSeek-R1 related workloads. “It’s a fairly expensive model to run inference on,” he said. “It has 671 billion parameters and you need to distribute it over multiple servers. And because the quality is higher, there’s generally more demand on the top end, which means you need more capacity.” Additionally, he noted that DeepSeek-R1 generally has longer-lived requests that can last two to three minutes. Tremendous user demand for DeepSeek-R1 is further driving the need for more infrastructure. To meet that demand, Together AI has rolled out a service it calls “reasoning clusters” that provision dedicated capacity, ranging from 128 to 2,000 chips, to run models at the best possible performance. How Together AI is helping organizations use reasoning AI There are a number of specific areas where Together AI is seeing usage of reasoning models. These include: Coding agents: Reasoning models help break down larger problems into steps. Reducing hallucinations: The reasoning process helps to verify the outputs of models, thus reducing hallucinations, which is important for applications where accuracy is crucial. Improving non-reasoning models: Customers are distilling and improving the quality of non-reasoning models. Enabling self-improvement: The use of reinforcement learning with reasoning models allows models to recursively self-improve without relying on large amounts of human-labeled data. Agentic AI is also driving increased demand for AI infrastructure  Together AI is also seeing increased infrastructure demand as its users embrace agentic AI. Prakash explained that agentic workflows, where a single user request results in thousands of API calls to complete a task, are putting more compute demand on Together AI’s infrastructure. To help support agentic AI workloads, Together AI recently has acquired CodeSandbox, whose technology provides lightweight, fast-booting virtual machines (VMs) to execute arbitrary, secure code within the Together AI cloud, where the language models also reside. This allows Together AI to reduce the latency between the agentic code and the models that need to be called, improving the performance of agentic workflows. Nvidia Blackwell is already having an impact All AI platforms are facing increased demands.  That’s one of the reasons why Nvidia keeps rolling out new silicon that provides more performance. Nvidia’s latest product chip is the Blackwell GPU, which is now being deployed at Together AI. Prakash said Nvidia Blackwell chips cost around 25% more than the previous generation, but provide 2X the performance. The GB 200 platform with Blackwell chips is particularly well-suited for training and inference of mixture of expert (MoE) models, which are trained across multiple InfiniBand-connected servers. He noted that Blackwell chips are also expected to provide a bigger performance boost for inference of larger models, compared to smaller models. The competitive landscape of agentic AI The market of AI infrastructure platforms is fiercely competitive.  Together AI faces competition from both established cloud providers and AI infrastructure startups. All the hyperscalers, including Microsoft, AWS and Google, have AI platforms. There is also an emerging category of AI-focussed players such as Groq and Samba Nova that are all aiming for a slice of the lucrative market. Together AI has a full-stack offering, including GPU infrastructure with software platform layers on top. This allows customers to easily build with open-source models or develop their own models on the Together AI platform. The company also has a focus on research developing optimizations and accelerated runtimes for both inference and training. “For instance, we serve the DeepSeek-R1 model at 85 tokens per second and Azure serves it at 7 tokens per second,” said Prakash. “There is a fairly widening gap in the performance and cost that we can provide to our customers.” source

By Jo Tunnicliff and Alex Kirkhope ( February 19, 2025, 3:13 PM GMT) — Donald Trump’s victory in the U.S. presidential election and recent subsequent inauguration on Jan. 21 has sparked a wave of reactions across business, including in the technology sector…. Law360 is on it, so you are, too. A Law360 subscription puts you at the center of fast-moving legal issues, trends and developments so you can act with speed and confidence. Over 200 articles are published daily across more than 60 topics, industries, practice areas and jurisdictions. A Law360 subscription includes features such as Daily newsletters Expert analysis Mobile app Advanced search Judge information Real-time alerts 450K+ searchable archived articles And more! Experience Law360 today with a free 7-day trial. source