Get ONMINE

Stay Ahead, Stay ONMINE

How a bot management file push crippled Cloudflare’s global network

As traffic moves through the core proxy, it applies each customer’s unique configuration and settings, from enforcing WAF rules and DDoS protection to routing traffic to the Developer Platform and R2. These configuration and policy rules are enforced by domain-specific modules. One of those modules, Bot Management, resulted in the outage. The Bot Management module […]

As traffic moves through the core proxy, it applies each customer’s unique configuration and settings, from enforcing WAF rules and DDoS protection to routing traffic to the Developer Platform and R2. These configuration and policy rules are enforced by domain-specific modules. One of those modules, Bot Management, resulted in the outage.

The Bot Management module uses a machine-learning model to assign bot scores to every request. The model relies on a feature configuration file, which is a collection of traits that help determine whether a request is automated. This feature file is refreshed every few minutes, allowing Cloudflare to react to new types of bots and new bot attacks.

A change in ClickHouse query behaviour that generates this file created large numbers of duplicate feature rows, increasing the size of the previously fixed-size feature configuration file, causing the bots module to trigger an error. As a result, Cloudflare’s core proxy began returning HTTP 5xx errors for any traffic relying on the bots module. Services like Workers KV and Access were also impacted, as they depend on the same core layer.

Once the underlying issue was identified, Cloudflare halted the generation and propagation of the bad feature file and manually inserted a known good file into the feature file distribution queue, and then forced a restart of its core proxy.

“This is like the butterfly effect in modern interconnected systems. A butterfly can flap its wings to fly, and its wing-flapping is the basis for the butterfly effect, a concept in chaos theory where a tiny change can have massive, unpredictable consequences. Same thing we are seeing here,” said Pareekh Jain, CEO at EIIRTrend & Pareekh Consulting.

Jain noted that such deep dependencies between security layers and AI modules for traffic routing are now common in large-scale cloud environments. Integrating powerful, dynamic modules directly into request processing pipelines means a flaw anywhere in the path can disrupt everything downstream.

Shape
Shape
Stay Ahead

Explore More Insights

More Articles

Stay ahead with more perspectives on cutting-edge power, infrastructure, energy,  bitcoin and AI solutions. Explore these articles to uncover strategies and insights shaping the future of industries.

Shape

Energy Secretary Strengthens Midwest Grid Reliability Heading into Winter Months

WASHINGTON—U.S. Secretary of Energy Chris Wright issued an emergency order to address critical grid reliability issues facing the Midwestern region of the United States heading into the cold winter months. The emergency order directs the Midcontinent Independent System Operator (MISO), in coordination with Consumers Energy, to ensure that the J.H. Campbell coal-fired power plant in West Olive, Michigan remains available for operation and to take every step to minimize costs for the American people. The Campbell Plant was scheduled to shut down on May 31, 2025 — 15 years before the end of its scheduled design life. “Because of the last administration’s dangerous energy subtraction policies targeting reliable and affordable energy sources, the United States continues to face an energy emergency,” said Energy Secretary Wright. “The Trump administration will keep taking action to reverse these energy subtraction policies, lowering energy costs and minimizing the risks of blackouts. Americans deserve access to affordable, reliable and secure energy regardless of whether the wind is blowing or the sun is shining, especially in dangerously cold weather.”  Since the Department of Energy’s (DOE) original order issued on May 23, the Campbell plant has proven critical to MISO’s operations, operating regularly during periods of high energy demand and low levels of intermittent energy production. A subsequent order was issued on August 20, 2025. As outlined in DOE’s Resource Adequacy Report, power outages could increase by 100 times in 2030 if the U.S. continues to take reliable power offline. The emergency conditions that led to the issuance of the original orders persist.MISO’s service area will continue to face emergency conditions both in the near and long term. Two recent winter studies (2024 – 2025 NERC Winter Reliability Assessment and the 2023 – 2024 NERC Winter Reliability Assessment) have assessed the MISO assessment area as an elevated risk, with the “potential

Read More »

South Sudan Says Crude Exports Back to Normal

South Sudan said it had resumed oil shipments after attacks on energy facilities in neighboring Sudan disrupted activity. “Operations in all oil fields in South Sudan have returned to a normal export,” Petroleum Ministry Undersecretary Deng Lual Wol told reporters Wednesday in the capital, Juba. “All crude exports from South Sudan are fully flowing to the export terminals in Port Sudan.” Oil companies operating in the two African countries earlier this week shuttered production after the assaults in Sudan, which is embroiled in a more than two-year civil war. Landlocked South Sudan uses pipelines to transport its crude to Red Sea terminals, from where it’s shipped to world markets. Dar Petroleum Operating Co. is producing 97,000 barrels per day following the brief shutdown, but will ramp that up to 150,000, Wol said. Greater Pioneer Operating Co.’s output is 40,000 daily barrels, and should rise to the normal level of 50,000, while Sudd Petroleum Operating Co. is pumping 13,000 barrels per day, down from 15,000 before disruption, he added. Bashayer Pipeline Co., which transports South Sudan’s Dar Blend oil to Sudan, said in a Nov. 15 notice seen by Bloomberg that it had initiated an emergency shutdown after its Al Jabalain processing plant and a power facility came under attack. Sudan’s state-owned Petrolines for Crude Oil Co. issued a Nov. 13 notice about a drone attack at the Heglig oil field, where Nile Blend is produced. It had issued a force majeure notice at 2B OPCO, an exploration and production company in which it has a 50 percent stake. What do you think? We’d love to hear from you, join the conversation on the Rigzone Energy Network. The Rigzone Energy Network is a new social experience created for you and all energy professionals to Speak Up about our industry, share knowledge, connect with

Read More »

Eni to Acquire 760 MW RE Assets in France from Neoen

Eni SpA said Tuesday it has entered into an agreement to buy a portfolio of already operational renewable energy projects totaling about 760 megawatts across France from Neoen. The transaction involves the transfer of 37 solar plants, 14 wind farms and one battery energy storage to Eni’s renewables arm Plenitude. The facilities produce around 1.1 terawatt hours of power annually, Italy’s state-backed Eni said in a press release. “The transaction represents one of the largest renewable energy deals completed in the French market in recent years and significantly contributes to Plenitude’s 2025 installed capacity targets”, Eni said. The parties have not disclosed the transaction price. Eni aims to reach over 5.5 gigawatts (GW) of installed renewable generation capacity this year, toward 10 GW by 2028 and 15 GW by 2030, according to a plan it announced February. As of the third quarter of 2025, it had 4.8 GW of installed renewable capacity, according to its quarterly report October 24. Eni plans to integrate the Neoen assets with its existing assets to “enable optimized operations and synergies”, Tuesday’s statement said. “The acquisition expands our presence in France, where we already serve around one million retail customers and where we are growing in both energy solutions and e-mobility markets”, said Plenitude chief executive Stefano Goberti. “Through this operation, we strengthen our integrated business model and accelerate progress toward achieving our strategic objectives”. Plenitude currently serves 10 million households and businesses across Europe, and aims to have over 11 million customers by 2028 and 15 million by 2030, Eni said. Paris-based Neoen said separately it would “continue to manage the plants for some years through the provision of asset management services to Plenitude”. Neoen said it would retain 1.1 GW of assets in operation or under construction in France including 754 MW of

Read More »

Monumental Completes Capital Raise to Fund More Production Restarts in NZ

Monumental Energy Corp said Tuesday it had completed the issuance of 16.2 million units for CAD 0.05 per unit in an oversubscribed non-brokered private placement, generating gross proceeds of CAD 810,000 ($580,000). Vancouver, Canada-based Monumental said in an online statement it would use net proceeds “to fund cost overruns on Copper Moki 1 oil and gas well, to fund the costs and expenses to formally enter into and fund additional workover projects with New Zealand Energy Corp. and L&M Energy and for general working capital purposes and corporate expenses”. “Each unit is comprised of one common share in the capital of the company and one transferable common share purchase warrant”, Toronto-listed Monumental said. “Each warrant entitles the holder thereof to purchase one additional common share of the company at a price of CAD 0.08 per share until November 18, 2028. “In connection with the private placement, the company paid in consideration of the services rendered by certain finders an aggregate cash commission of CAD 38,850 and issued an aggregate of 777,000 non-transferable common share purchase warrants. Each finder warrant entitles the holder thereof to purchase one additional common share of the company at the issuer price until November 18, 2028”. Last month Monumental said it has agreed to fund New Zealand Energy Corp’s (NZEC) share of workover costs to restart flows at several wells in the Waihapa/Ngaere field in the onshore Taranaki basin. “These workovers will follow the same royalty structure as that established for the successful Copper Moki programs, whereas Monumental will earn a 25 percent royalty on NZEC’s production share after full recovery of its capital investment, which will be repaid from 75 percent of NZEC’s net revenue interest”, Monumental, a shareholder in NZEC, said in a press release October 15. L&M Energy will shoulder the remaining investment as NZEC’s

Read More »

US Risks Winter Blackouts on Data Center Demand

Rising electricity demand from data centers is raising the risk of blackouts across a wide swath of the US during extreme conditions this winter, according to the regulatory body overseeing grid stability.  Power consumption has grown 20 gigawatts from the previous winter, the North American Electric Reliability Corp. said Tuesday in its winter assessment.  A gigawatt is the typical size of a nuclear power reactor. Supply hasn’t kept up.  As as result, a repeat of severe winter storms in North America that unleash a polar vortex, of which there have been several in recent years, could trigger energy shortfalls across the US from the Northwest to Texas to the Carolinas. All regions have adequate resources in normal conditions. “Data centers are a main contributor to load growth in those areas where demand has risen substantially since last winter,” Mark Olson, manager of the reliability assessment, said in an emailed statement.  America’s power grid has been facing rising blackout risks for years as aging infrastructure is increasingly stressed by severe storms and wildfires. Now the data center boom, driven by the spread of artificial intelligence, is adding to the strain by supercharging US electricity growth after being stagnant for two decades.  Winter is especially risky because solar generation is available for fewer hours and battery operations may be affected. Gas supplies, meantime, could drop off because of freeze-offs or pipeline constraints. The areas designated by NERC as having elevated risks of shortfall shifted from the previous winter to include the US southeast and parts of the West, including Washington and Oregon.  The Texas grid continues to be highlighted after cascading failures in February 2021 left millions of people without power for days and resulted in more than 200 deaths. New England also continues to face elevated risks on potential natural gas pipeline

Read More »

US Energy Majors among Potential Lukoil Bidders

Exxon Mobil Corp., Chevron Corp., Abu Dhabi National Oil Co. as well as US private equity giant Carlyle Group are among companies interested in Lukoil PJSC’s international assets, a sale hastened by US sanctions due to kick in next month. Suitors are lining up to look at the various parts of Russian energy giant’s sprawling international business, with some potential buyers only interested in specific assets, according to people familiar with the situation. But one potential problem is that Lukoil favors selling the assets as a single package ahead of sanctions due to take effect Dec. 13, one of the people said. This raises the possibility of a two-step process in which one buyer – such as a financial firm – acquires all of Lukoil’s non-Russian assets and then resells them piecemeal over time. A key detail in the process is that the Trump administration would prefer that Lukoil’s global assets are taken over by a US entity, a fact that may limit the pool of potential buyers, people with knowledge of the matter said. A spokesperson for the US Treasury didn’t immediately respond to a request for comment.  Lukoil had previously agreed to sell the whole international business to Gunvor Group – a deal that was then dramatically blocked by the US.  Exxon and Chevron are exploring Lukoil’s stake in the West Qurna 2 field in Iraq, said two of the people, who declined to be identified because the talks are private. Meanwhile Adnoc is looking at various Lukoil assets, with the Russian firm’s natural gas operations in Uzbekistan potentially of most interest, according to people with knowledge of the situation. Spokespeople for Chevron, Exxon, Carlyle and Adnoc’s international investment unit XRG all declined to comment. Lukoil didn’t respond to a request to do so. What do you think? We’d love to

Read More »

Nvidia’s first exascale system is the 4th fastest supercomputer in the world

The world’s fourth exascale supercomputer has arrived, pitting Nvidia’s proprietary chip technologies against the x86 systems that have dominated supercomputing for decades. For the 66th edition of the TOP500, El Capitan holds steady at No. 1 while JUPITER Booster becomes the fourth exascale system on the list. The JUPITER Booster supercomputer, installed in Germany, uses Nvidia CPUs and GPUs and delivers a peak performance of exactly 1 exaflop, according to the November TOP500 list of supercomputers, released on Monday. The exaflop measurement is considered a major milestone in pushing computing performance to the limits. Today’s computers are typically measured in gigaflops and teraflops—and an exaflop translates to 1 billion gigaflops. Nvidia’s GPUs dominate AI servers installed in data centers as computing shifts to AI. As part of this shift, AI servers with Nvidia’s ARM-based Grace CPUs are emerging as a high-performance alternative to x86 chips. JUPITER is the fourth-fastest supercomputer in the world, behind three systems with x86 chips from AMD and Intel, according to TOP500. The top three supercomputers on the TOP500 list are in the U.S. and owned by the U.S. Department of Energy. The top two supercomputers—the 1.8-exaflop El Capitan at Lawrence Livermore National Laboratory and the 1.35-exaflop Frontier at Oak Ridge National Laboratory—use AMD CPUs and GPUs. The third-ranked 1.01-exaflop Aurora at Argonne National Laboratory uses Intel CPUs and GPUs. Intel scrapped its GPU roadmap after the release of Aurora and is now restructuring operations. The JUPITER Booster, which was assembled by France-based Eviden, has Nvidia’s GH200 superchip, which links two Nvidia Hopper GPUs with CPUs based on ARM designs. The CPU and GPU are connected via Nvidia’s proprietary NVLink interconnect, which is based on InfiniBand and provides bandwidth of up to 900 gigabytes per second. JUPITER first entered the Top500 list at 793 petaflops, but

Read More »

Samsung’s 60% memory price hike signals higher data center costs for enterprises

Industry-wide price surge driven by AI Samsung is not alone in raising prices. In October, TrendForce reported that Samsung and SK Hynix raised DRAM and NAND flash prices by up to 30% for Q4. Similarly, SK Hynix said during its October earnings call that its HBM, DRAM, and NAND capacity is “essentially sold out” for 2026, with the company posting record quarterly operating profit exceeding $8 billion, driven by surging AI demand. Industry analysts attributed the price increases to manufacturers redirecting production capacity. HBM production for AI accelerators consumes three times the wafer capacity of standard DRAM, according to a TrendForce report, citing remarks from Micron’s Chief Business Officer. After two years of oversupply, memory inventories have dropped to approximately eight weeks from over 30 weeks in early 2023. “The memory industry is tightening faster than expected as AI server demand for HBM, DDR5, and enterprise SSDs far outpaces supply growth,” said Manish Rawat, semiconductor analyst at TechInsights. “Even with new fab capacity coming online, much of it is dedicated to HBM, leaving conventional DRAM and NAND undersupplied. Memory is shifting from a cyclical commodity to a strategic bottleneck where suppliers can confidently enforce price discipline.” This newfound pricing power was evident in Samsung’s approach to contract negotiations. “Samsung’s delayed pricing announcement signals tough behind-the-scenes negotiations, with Samsung ultimately securing the aggressive hike it wanted,” Rawat said. “The move reflects a clear power shift toward chipmakers: inventories are normalized, supply is tight, and AI demand is unavoidable, leaving buyers with little room to negotiate.” Charlie Dai, VP and principal analyst at Forrester, said the 60% increase “signals confidence in sustained AI infrastructure growth and underscores memory’s strategic role as the bottleneck in accelerated computing.” Servers to cost 10-25% more For enterprises building AI infrastructure, these supply dynamics translate directly into

Read More »

Arista, Palo Alto bolster AI data center security

“Based on this inspection, the NGFW creates a comprehensive, application-aware security policy. It then instructs the Arista fabric to enforce that policy at wire speed for all subsequent, similar flows,” Kotamraju wrote. “This ‘inspect-once, enforce-many’ model delivers granular zero trust security without the performance bottlenecks of hairpinning all traffic through a firewall or forcing a costly, disruptive network redesign.” The second capability is a dynamic quarantine feature that enables the Palo Alto NGFWs to identify evasive threats using Cloud-Delivered Security Services (CDSS). “These services, such as Advanced WildFire for zero-day malware and Advanced Threat Prevention for unknown exploits, leverage global threat intelligence to detect and block attacks that traditional security misses,” Kotamraju wrote. The Arista fabric can intelligently offload trusted, high-bandwidth “elephant flows” from the firewall after inspection, freeing it to focus on high-risk traffic. When a threat is detected, the NGFW signals Arista CloudVision, which programs the network switches to automatically quarantine the compromised workload at hardware line-rate, according to Kotamraju: “This immediate response halts the lateral spread of a threat without creating a performance bottleneck or requiring manual intervention.” The third feature is unified policy orchestration, where Palo Alto Networks’ management plane centralizes zone-based and microperimeter policies, and CloudVision MSS responds with the offload and enforcement of Arista switches. “This treats the entire geo-distributed network as a single logical switch, allowing workloads to be migrated freely across cloud networks and security domains,” Srikanta and Barbieri wrote. Lastly, the Arista Validated Design (AVD) data models enable network-as-a-code, integrating with CI/CD pipelines. AVDs can also be generated by Arista’s AVA (Autonomous Virtual Assist) AI agents that incorporate best practices, testing, guardrails, and generated configurations. “Our integration directly resolves this conflict by creating a clean architectural separation that decouples the network fabric from security policy. This allows the NetOps team (managing the Arista

Read More »

AMD outlines ambitious plan for AI-driven data centers

“There are very beefy workloads that you must have that performance for to run the enterprise,” he said. “The Fortune 500 mainstream enterprise customers are now … adopting Epyc faster than anyone. We’ve seen a 3x adoption this year. And what that does is drives back to the on-prem enterprise adoption, so that the hybrid multi-cloud is end-to-end on Epyc.” One of the key focus areas for AMD’s Epyc strategy has been our ecosystem build out. It has almost 180 platforms, from racks to blades to towers to edge devices, and 3,000 solutions in the market on top of those platforms. One of the areas where AMD pushes into the enterprise is what it calls industry or vertical workloads. “These are the workloads that drive the end business. So in semiconductors, that’s telco, it’s the network, and the goal there is to accelerate those workloads and either driving more throughput or drive faster time to market or faster time to results. And we almost double our competition in terms of faster time to results,” said McNamara. And it’s paying off. McNamara noted that over 60% of the Fortune 100 are using AMD, and that’s growing quarterly. “We track that very, very closely,” he said. The other question is are they getting new customer acquisitions, customers with Epyc for the first time? “We’ve doubled that year on year.” AMD didn’t just brag, it laid out a road map for the next two years, and 2026 is going to be a very busy year. That will be the year that new CPUs, both client and server, built on the Zen 6 architecture begin to appear. On the server side, that means the Venice generation of Epyc server processors. Zen 6 processors will be built on 2 nanometer design generated by (you guessed

Read More »

Building the Regional Edge: DartPoints CEO Scott Willis on High-Density AI Workloads in Non-Tier-One Markets

When DartPoints CEO Scott Willis took the stage on “the Distributed Edge” panel at the 2025 Data Center Frontier Trends Summit, his message resonated across a room full of developers, operators, and hyperscale strategists: the future of AI infrastructure will be built far beyond the nation’s tier-one metros. On the latest episode of the Data Center Frontier Show, Willis expands on that thesis, mapping out how DartPoints has positioned itself for a moment when digital infrastructure inevitably becomes more distributed, and why that moment has now arrived. DartPoints’ strategy centers on what Willis calls the “regional edge”—markets in the Midwest, Southeast, and South Central regions that sit outside traditional cloud hubs but are increasingly essential to the evolving AI economy. These are not tower-edge micro-nodes, nor hyperscale mega-campuses. Instead, they are regional data centers designed to serve enterprises with colocation, cloud, hybrid cloud, multi-tenant cloud, DRaaS, and backup workloads, while increasingly accommodating the AI-driven use cases shaping the next phase of digital infrastructure. As inference expands and latency-sensitive applications proliferate, Willis sees the industry’s momentum bending toward the very markets DartPoints has spent years cultivating. Interconnection as Foundation for Regional AI Growth A key part of the company’s differentiation is its interconnection strategy. Every DartPoints facility is built to operate as a deeply interconnected environment, drawing in all available carriers within a market and stitching sites together through a regional fiber fabric. Willis describes fiber as the “nervous system” of the modern data center, and for DartPoints that means creating an interconnection model robust enough to support a mix of enterprise cloud, multi-site disaster recovery, and emerging AI inference workloads. The company is already hosting latency-sensitive deployments in select facilities—particularly inference AI and specialized healthcare applications—and Willis expects such deployments to expand significantly as regional AI architectures become more widely

Read More »

Key takeaways from Cisco Partner Summit

Brian Ortbals, senior vice president from World Wide Technology, which is one of Cisco’s biggest and most important partners stated: “Cisco engaged partners early in the process and took our feedback along the way. We believe now is the right time for these changes as it will enable us to capitalize on the changes in the market.” The reality is, the more successful its more-than-half-a-million partners are, the more successful Cisco will be. Platform approach is coming together When Jeetu Patel took the reigns as chief product officer, one of his goals was to make the Cisco portfolio a “force multiple.” Patel has stated repeatedly that, historically, Cisco acted more as a technology holding company with good products in networking, security, collaboration, data center and other areas. In this case, product breadth was not an advantage, as everything must be sold as “best of breed,” which is a tough ask of the salesforce and partner community. Since then, there have been many examples of the coming together of the portfolio to create products that leverage the breadth of the platform. The latest is the Unified Edge appliance, an all-in-one solution that brings together compute, networking, storage and security. Cisco has been aggressive with AI products in the data center, and Cisco Unified Edge compliments that work with a device designed to bring AI to edge locations. This is ideally suited for retail, manufacturing, healthcare, factories and other industries where it’s more cost effecting and performative to run AI where the data lives.

Read More »

Microsoft will invest $80B in AI data centers in fiscal 2025

And Microsoft isn’t the only one that is ramping up its investments into AI-enabled data centers. Rival cloud service providers are all investing in either upgrading or opening new data centers to capture a larger chunk of business from developers and users of large language models (LLMs).  In a report published in October 2024, Bloomberg Intelligence estimated that demand for generative AI would push Microsoft, AWS, Google, Oracle, Meta, and Apple would between them devote $200 billion to capex in 2025, up from $110 billion in 2023. Microsoft is one of the biggest spenders, followed closely by Google and AWS, Bloomberg Intelligence said. Its estimate of Microsoft’s capital spending on AI, at $62.4 billion for calendar 2025, is lower than Smith’s claim that the company will invest $80 billion in the fiscal year to June 30, 2025. Both figures, though, are way higher than Microsoft’s 2020 capital expenditure of “just” $17.6 billion. The majority of the increased spending is tied to cloud services and the expansion of AI infrastructure needed to provide compute capacity for OpenAI workloads. Separately, last October Amazon CEO Andy Jassy said his company planned total capex spend of $75 billion in 2024 and even more in 2025, with much of it going to AWS, its cloud computing division.

Read More »

John Deere unveils more autonomous farm machines to address skill labor shortage

Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn More Self-driving tractors might be the path to self-driving cars. John Deere has revealed a new line of autonomous machines and tech across agriculture, construction and commercial landscaping. The Moline, Illinois-based John Deere has been in business for 187 years, yet it’s been a regular as a non-tech company showing off technology at the big tech trade show in Las Vegas and is back at CES 2025 with more autonomous tractors and other vehicles. This is not something we usually cover, but John Deere has a lot of data that is interesting in the big picture of tech. The message from the company is that there aren’t enough skilled farm laborers to do the work that its customers need. It’s been a challenge for most of the last two decades, said Jahmy Hindman, CTO at John Deere, in a briefing. Much of the tech will come this fall and after that. He noted that the average farmer in the U.S. is over 58 and works 12 to 18 hours a day to grow food for us. And he said the American Farm Bureau Federation estimates there are roughly 2.4 million farm jobs that need to be filled annually; and the agricultural work force continues to shrink. (This is my hint to the anti-immigration crowd). John Deere’s autonomous 9RX Tractor. Farmers can oversee it using an app. While each of these industries experiences their own set of challenges, a commonality across all is skilled labor availability. In construction, about 80% percent of contractors struggle to find skilled labor. And in commercial landscaping, 86% of landscaping business owners can’t find labor to fill open positions, he said. “They have to figure out how to do

Read More »

2025 playbook for enterprise AI success, from agents to evals

Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn More 2025 is poised to be a pivotal year for enterprise AI. The past year has seen rapid innovation, and this year will see the same. This has made it more critical than ever to revisit your AI strategy to stay competitive and create value for your customers. From scaling AI agents to optimizing costs, here are the five critical areas enterprises should prioritize for their AI strategy this year. 1. Agents: the next generation of automation AI agents are no longer theoretical. In 2025, they’re indispensable tools for enterprises looking to streamline operations and enhance customer interactions. Unlike traditional software, agents powered by large language models (LLMs) can make nuanced decisions, navigate complex multi-step tasks, and integrate seamlessly with tools and APIs. At the start of 2024, agents were not ready for prime time, making frustrating mistakes like hallucinating URLs. They started getting better as frontier large language models themselves improved. “Let me put it this way,” said Sam Witteveen, cofounder of Red Dragon, a company that develops agents for companies, and that recently reviewed the 48 agents it built last year. “Interestingly, the ones that we built at the start of the year, a lot of those worked way better at the end of the year just because the models got better.” Witteveen shared this in the video podcast we filmed to discuss these five big trends in detail. Models are getting better and hallucinating less, and they’re also being trained to do agentic tasks. Another feature that the model providers are researching is a way to use the LLM as a judge, and as models get cheaper (something we’ll cover below), companies can use three or more models to

Read More »

OpenAI’s red teaming innovations define new essentials for security leaders in the AI era

Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn More OpenAI has taken a more aggressive approach to red teaming than its AI competitors, demonstrating its security teams’ advanced capabilities in two areas: multi-step reinforcement and external red teaming. OpenAI recently released two papers that set a new competitive standard for improving the quality, reliability and safety of AI models in these two techniques and more. The first paper, “OpenAI’s Approach to External Red Teaming for AI Models and Systems,” reports that specialized teams outside the company have proven effective in uncovering vulnerabilities that might otherwise have made it into a released model because in-house testing techniques may have missed them. In the second paper, “Diverse and Effective Red Teaming with Auto-Generated Rewards and Multi-Step Reinforcement Learning,” OpenAI introduces an automated framework that relies on iterative reinforcement learning to generate a broad spectrum of novel, wide-ranging attacks. Going all-in on red teaming pays practical, competitive dividends It’s encouraging to see competitive intensity in red teaming growing among AI companies. When Anthropic released its AI red team guidelines in June of last year, it joined AI providers including Google, Microsoft, Nvidia, OpenAI, and even the U.S.’s National Institute of Standards and Technology (NIST), which all had released red teaming frameworks. Investing heavily in red teaming yields tangible benefits for security leaders in any organization. OpenAI’s paper on external red teaming provides a detailed analysis of how the company strives to create specialized external teams that include cybersecurity and subject matter experts. The goal is to see if knowledgeable external teams can defeat models’ security perimeters and find gaps in their security, biases and controls that prompt-based testing couldn’t find. What makes OpenAI’s recent papers noteworthy is how well they define using human-in-the-middle

Read More »
More Articles