New Jersey BPU eyes potential for new nuclear

Stay Ahead, Stay ONMINE

New Jersey BPU eyes potential for new nuclear

The New Jersey Board of Public Utilities on Tuesday issued a request for information to explore whether new nuclear power plants could advance the state’s goals on electricity affordability, resource adequacy and clean energy. The RFI comes as modeling for New Jersey’s upcoming Energy Master Plan indicates New Jersey will need up to 10 GW […]

The RFI comes as modeling for New Jersey’s upcoming Energy Master Plan indicates New Jersey will need up to 10 GW of clean, firm capacity to maintain system reliability after 2035, partly in response to growing data center demand forecasts, the BPU said. New Jersey aims to have 100% clean electricity by 2035.

“As we work to push PJM to improve their interconnection queue to allow more resources like solar and storage to be built in the short-term, expanding our nuclear fleet offers the Garden State an opportunity to add new generation to our resource mix, improving reliability and affordability for ratepayers in the long-term,” Christine Guhl-Sadovy, BPU president, said in a press release.

New Jersey has three nuclear power reactors: the Salem 1 and 2 units, which total 2,285 MW, and the 1,172-MW Hope Creek unit. PSEG Power owns the Hope Creek plant and 57% of the Salem capacity. The rest of the Salem plant is owned by Constellation Energy. The reactors generate 40% of New Jersey’s electricity and account for 85% of the state’s emission-free generation, according to the BPU.

In the RFI, BPU staff is seeking answers to a range of questions, including how nuclear electricity production can address the needs of large loads such as AI data centers and how ratepayers can be protected from the effects of using existing nuclear power plants for data centers.

Agency staff also want to know whether New Jersey should provide financial assistance or incentives for new nuclear generating capacity and how incentive programs could be designed to support the state’s economy while protecting ratepayers.

“What environmental impacts and benefits should be considered as the state contemplates the development of new nuclear electric generating capacity in New Jersey?” BPU staff asked in the RFI.

Responses to the RFI are due on June 23.

Stay Ahead

Explore More Insights

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.

IBM introduces new generation of LinuxOne AI mainframe

In addition to generative AI applications, new multiple model AI approaches are engineered to enhance prediction and accuracy in many industry use cases like advanced fraud detection, image processing and retail automation, according to IBM. LinuxONE Emperor 5 also comes with advanced security features specifically designed for the AI threat

Juniper extends Mist AI observability, performance management capabilities

“Unlike traditional solutions for digital twinning and synthetic testing, Marvis Minis don’t require manual configuration or any additional hardware or software. They are digital experience twins, now client-to-cloud available on all Juniper full-stack devices,” according to a data sheet from Juniper. “Marvis Minis are always on and constantly ingesting user

StorONE and Phison partner for AI-native on-premises storage platform

The idea is to let customers to train the largest language models — up to several billion parameters — on-premises with less hardware required while offering a built-in conversational chatbot that helps users manage, query, and optimize storage configurations and performance in plain English. “In today’s AI-first world, infrastructure must evolve beyond raw

IBM wrangles AI agents to work across complex enterprise environments

In addition, the new Agent Catalog in watsonx Orchestrate can simplify access to more than 150 agents and pre-built tools from IBM and its partners, which include Box, MasterCard, Oracle, Salesforce, ServiceNow, and Symplistic.ai. IBM is also rolling out an agent builder tool in June that will let customers build their own agents in less

Harbour Energy reports ‘strong start to the year’ despite job cuts

Following news that it would cut 250 jobs from its onshore Aberdeen workforce, Harbour Energy has reported a “strong start to the year”. The UK’s largest oil and gas producer reiterated points about the UK’s challenging fiscal regime being at the heart of the decision to slash its north-east headcount by 25%, adding that “significantly lower investment levels” played a part in the decision. Business commentators said that the latest Harbour Energy redundancies served as a “devastating blow” for those impacted and that the news was “just the tip of the iceberg, unless the government changes course”. Despite this, chief executive Linda Cook announced news of “a strong start to the year” to shareholders on Thursday morning. In the first quarter of the year, Harbour Energy production levels averaged out at 500 thousand barrels of oil equivalent per day (kboepd). This, alongside “improved European gas price realisations and lower unit costs”, has resulted in revenues of $2.8 billion and a free cash flow of $700 million. Cook added: “Recent market volatility reinforces the benefits of our diverse portfolio and our prudent approach to risk management. ” Following the takeover of Wintershall Dea last year, Norway is now Harbour Energy’s largest producing region. However, in the first quarter it brought new wells on stream in the UK, as well as Argentina, Egypt and Germany. The UK is still the North Sea player’s second largest producing area, falling just shy of its Nordic neighbour. Harbour produced 180 kboepd in the first period of 2025 while producing 165 kboepd in the UK. These figures mark a marginal uptick in production in the UK since last year’s reporting of 161 kboepd. In its latest trading update, the firm announced that its production guidance for the year had increased slightly, moving from 450-475 kboepd to

Energy Secretary Wright Testifies Before House Appropriations Subcommittee on FY2026 Budget Request

WASHINGTON — U.S. Secretary of Energy Chris Wright testified today before the House Committee on Appropriations, Subcommittee on Energy and Water Development, outlining the Department of Energy’s Fiscal Year 2026 budget request. The FY2026 Budget reflects President Trump’s directive to unleash American energy, eliminate wasteful spending and refocus the Department on its core mission. It brings non-defense discretionary spending to its leanest level since 2017 and eliminates over $15 billion in Green New Scam funding that supports more expensive, less reliable energy sources. For more details, view the budget toplines here. Secretary Wright’s opening remarks: Chairman Fleischmann, Ranking Member Kaptur, and Members of the Committee, it is an honor to appear before you and this Committee today to discuss the President’s Fiscal Year 2026 Budget request for the Department of Energy. I’m especially honored to be at my first hearing before you as U.S. Secretary of Energy. I want to commend this Committee for its longstanding commitment to energy policy and to the mission of the Department of Energy. Energy is the backbone of civilization. It is the essential catalyst of human progress— enabling everything that we do, from the lights in our home, the process heat in our factories and the innovation in our national laboratories. I’ve dedicated my life to increasing access to energy, and I’m thrilled to carry my work forward at the Department of Energy at this pivotal moment in our history. My priorities for the Department are clear — to unleash a golden era of American energy dominance, strengthen our national security, and lead the world in innovation. A reliable and abundant energy supply is the foundation of a strong and prosperous nation– it drives our economy, safeguards our freedoms and fuels breakthroughs that improve our lives. When America leads in energy, we lead in prosperity, security and human

ScottishPower receives £600m for UK grid upgrades

The National Wealth Fund has handed out £600 million to ScottishPower to finance further grid upgrades in the UK. The Iberdrola subsidiary has said that the investment will help support the creation of “hundreds of jobs” while supporting the country on its journey to net zero. This funding will support seven of ScottishPower’s priority transmission grid upgrade projects, including Eastern Green Link 1 and 4, which aim to build interconnectors between Scotland and England. ScottishPower claims that the projects being financed by the National Wealth Fund will support increased renewable electricity into the UK grid. Keith Anderson, chief executive of ScottishPower, said: “Today’s announcement is a welcome step forward in the delivery of the Government’s Clean Power 2030 pathway. “Working together to drive forward these critical investments is an important catalyst for economic growth, as we make progress in bringing more renewables onto the system to meet the increasing demand for electrification.” © Supplied by ScottishPowerScottishPower’s Carland Cross wind farm and grid infrastructure. The investment forms part of a £1.35 billion financing package, led by Bank of America as Sole Debt Arranger and including Bank of America, BankInter, BNP Paribas, Caixabank, Lloyds Bank, NatWest and Banco Sabadell as lenders. ScottishPower argued that the amount of cash needed to deliver on government plans for green energy roll-out is “significant”. Recent data from the National Energy System Operator (NESO) suggested that up to £60 billion of investment is required between now and 2030 to support the delivery of a clean power system. NESO has previously identified Eastern Green Link 1 as a “crucial project”, ScottishPower said. Chancellor of the exchequer, Rachel Reeves, commented: “Upgrading our energy infrastructure is good news for businesses, households, and the economy. “It will bring down bills, put more money in working people’s pockets and enable businesses to

What training for a marathon has taught me about wellbeing

I ran in the London Marathon this year. That’s 26.2 miles with over 50,000 people on one of the most well-known marathon courses in the world. Taking part led to me celebrating another checkmark on my lifetime bucket list. They say persistence pays off, and after six times on the ballot, I found out on 3 February that my place was confirmed. That left me just 12 weeks to train, and what I’ve learned in that time is how our bodies and minds are so much stronger than we think. Marathon training has pushed me to limits I didn’t know I had and then showed me I could push beyond them. I discovered the journey was not just about running, the most valuable lessons were found in the training and the quiet determination that will get me to the start line. © Supplied by Space SolutionsSophie Taysom, Space Solutions. Exercise really does make us happy We know stress and anxiety levels are at record highs, and we also know that exercise is one of the best ways to manage our mental wellbeing. Research shows that after just 15 minutes and nine seconds of exercise, people start to feel a positive uplift. Now, a marathon is a bit longer than 15 minutes, but it’s a starting point. Each run releases a cascade of neurochemicals—endorphin boost, increased serotonin, and cortisol reduction—all nature’s own stress relievers. During the longer weekly runs, I felt my anxieties dissolve with each mile. A simple escapism to get me away from the daily mental stimulations. I was getting outside, away from home and work, and was given a feeling of freedom and control in a world which at times can feel overwhelming. Between constant notifications, family responsibilities and work, finding time for myself initially felt selfish. But

ISO New England prepares for 11% rise in annual electricity consumption

The electrification of heating systems and the adoption of electric vehicles are expected to drive New England annual electricity consumption to rise 11% over the next decade, according to the region’s grid operator. “The ISO is forecasting steady growth in net annual energy use as state policy goals for carbon emissions reductions drive the increased electrification of heating systems and transportation in the region,” the regional grid operator said in a Thursday blog post about its 2025–2034 Forecast Report of Capacity, Energy, Loads, and Transmission. Net annual energy usage in the region has trended down for the last two decades, the ISO said, “mainly due to an increase in energy efficiency from advanced cooling and heating technologies, energy-efficient appliances and lighting, and the increased prevalence of [behind-the-meter solar PV] generation. However, ISO New England is predicting that trend will reverse in the next decade.” The regional grid serves Connecticut, Maine, Massachusetts, New Hampshire, Rhode Island and Vermont. The report indicates net annual energy use across the region is expected to grow by 1.2% annually through 2034. Behind-the-meter solar will reduce consumption of grid-supplied electricity by 5,555 GWh this year, rising to 11,487 GWh in 2034, the ISO said. Electric vehicles are expected to account for 8,735 GWh of energy use in 2034, while heating electrification is expected to account for 8,049 GWh that year, the grid operator said. Summer net peak demand is also expected to rise by 0.9% annually across the next 10 years. Behind-the-meter solar will reduce the summer peak by an average of 1,870 MW, the grid operator said. “Although in past years the summer peak typically occurred in the afternoon, widespread adoption of BTM PV has effectively pushed the peak toward evening,” ISO New England said. “This means that even as more BTM PV comes on

Oil Drops on Powell Rate Comments

Oil slumped after Federal Reserve Chair Jerome Powell said officials aren’t in a hurry to adjust rates and as tensions between the US and China linger ahead of trade negotiations. West Texas Intermediate futures dipped 1.7% to settle near $58 a barrel as the Fed outlook and trade conflict weighed on the prospects for economic growth and energy demand. President Donald Trump said he’s unwilling to preemptively lower tariffs on China, the world’s largest crude importer, to kickstart substantive trade talks. The comments come a day before Treasury Secretary Scott Bessent and Trade Representative Jamieson Greer are set to meet with Chinese officials. In the US, the Federal Reserve held interest rates steady in their meeting Wednesday, and Powell said tariffs could lead to higher inflation and unemployment. “There is a fear that the trade negotiations in Switzerland with China could backfire and turn into a demand destruction event,” said Robert Yawger, director of the energy futures division at Mizuho Securities USA. Market sentiment that a rate-cut isn’t coming anytime soon is also weighing on prices, he added. Oil has trended lower since late January due to escalating trade frictions and plans by OPEC+ to keep boosting idled supply. Also weighing on crude, Trump said the US would stop its bombing campaign against Houthis in Yemen after a ceasefire was facilitated by Oman. That followed earlier comments from Vice President JD Vance that a nuclear deal with Iran could see the OPEC member reintegrated into the global economy. The decline in crude prices will likely lead to falling American shale output, according to Diamondback Energy Inc., the largest US independent oil producer in the Permian Basin. In another sign of the hit to output, the Energy Information Administration cut its forecast for US crude production this year for a second

Liquid cooling technologies: reducing data center environmental impact

“Highly optimized cold-plate or one-phase immersion cooling technologies can perform on par with two-phase immersion, making all three liquid-cooling technologies desirable options,” the researchers wrote. Factors to consider There are numerous factors to consider when adopting liquid cooling technologies, according to Microsoft’s researchers. First, they advise performing a full environmental, health, and safety analysis, and end-to-end life cycle impact analysis. “Analyzing the full data center ecosystem to include systems interactions across software, chip, server, rack, tank, and cooling fluids allows decision makers to understand where savings in environmental impacts can be made,” they wrote. It is also important to engage with fluid vendors and regulators early, to understand chemical composition, disposal methods, and compliance risks. And associated socioeconomic, community, and business impacts are equally critical to assess. More specific environmental considerations include ozone depletion and global warming potential; the researchers emphasized that operators should only use fluids with low to zero ozone depletion potential (ODP) values, and not hydrofluorocarbons or carbon dioxide. It is also critical to analyze a fluid’s viscosity (thickness or stickiness), flammability, and overall volatility. And operators should only use fluids with minimal bioaccumulation (the buildup of chemicals in lifeforms, typically in fish) and terrestrial and aquatic toxicity. Finally, once up and running, data center operators should monitor server lifespan and failure rates, tracking performance uptime and adjusting IT refresh rates accordingly.

Cisco unveils prototype quantum networking chip

Clock synchronization allows for coordinated time-dependent communications between end points that might be cloud databases or in large global databases that could be sitting across the country or across the world, he said. “We saw recently when we were visiting Lawrence Berkeley Labs where they have all of these data sources such as radio telescopes, optical telescopes, satellites, the James Webb platform. All of these end points are taking snapshots of a piece of space, and they need to synchronize those snapshots to the picosecond level, because you want to detect things like meteorites, something that is moving faster than the rotational speed of planet Earth. So the only way you can detect that quickly is if you synchronize these snapshots at the picosecond level,” Pandey said. For security use cases, the chip can ensure that if an eavesdropper tries to intercept the quantum signals carrying the key, they will likely disturb the state of the qubits, and this disturbance can be detected by the legitimate communicating parties and the link will be dropped, protecting the sender’s data. This feature is typically implemented in a Quantum Key Distribution system. Location information can serve as a critical credential for systems to authenticate control access, Pandey said. The prototype quantum entanglement chip is just part of the research Cisco is doing to accelerate practical quantum computing and the development of future quantum data centers. The quantum data center that Cisco envisions would have the capability to execute numerous quantum circuits, feature dynamic network interconnection, and utilize various entanglement generation protocols. The idea is to build a network connecting a large number of smaller processors in a controlled environment, the data center warehouse, and provide them as a service to a larger user base, according to Cisco. The challenges for quantum data center network fabric

Zyxel launches 100GbE switch for enterprise networks

Port specifications include: 48 SFP28 ports supporting dual-rate 10GbE/25GbE connectivity 8 QSFP28 ports supporting 100GbE connections Console port for direct management access Layer 3 routing capabilities include static routing with support for access control lists (ACLs) and VLAN segmentation. The switch implements IEEE 802.1Q VLAN tagging, port isolation, and port mirroring for traffic analysis. For link aggregation, the switch supports IEEE 802.3ad for increased throughput and redundancy between switches or servers. Target applications and use cases The CX4800-56F targets multiple deployment scenarios where high-capacity backbone connectivity and flexible port configurations are required. “This will be for service providers initially or large deployments where they need a high capacity backbone to deliver a primarily 10G access layer to the end point,” explains Nguyen. “Now with Wi-Fi 7, more 10G/25G capable POE switches are being powered up and need interconnectivity without the bottleneck. We see this for data centers, campus, MDU (Multi-Dwelling Unit) buildings or community deployments.” Management is handled through Zyxel’s NebulaFlex Pro technology, which supports both standalone configuration and cloud management via the Nebula Control Center (NCC). The switch includes a one-year professional pack license providing IGMP technology and network analytics features. The SFP28 ports maintain backward compatibility between 10G and 25G standards, enabling phased migration paths for organizations transitioning between these speeds.

Engineers rush to master new skills for AI-driven data centers

According to the Uptime Institute survey, 57% of data centers are increasing salary spending. Data center job roles that saw the highest increases were in operations management – 49% of data center operators said they saw highest increases in this category – followed by junior and mid-level operations staff at 45%, and senior management and strategy at 35%. Other job categories that saw salary growth were electrical, at 32% and mechanical, at 23%. Organizations are also paying premiums on top of salaries for particular skills and certifications. Foote Partners tracks pay premiums for more than 1,300 certified and non-certified skills for IT jobs in general. The company doesn’t segment the data based on whether the jobs themselves are data center jobs, but it does track 60 skills and certifications related to data center management, including skills such as storage area networking, LAN, and AIOps, and 24 data center-related certificates from Cisco, Juniper, VMware and other organizations. “Five of the eight data center-related skills recording market value gains in cash pay premiums in the last twelve months are all AI-related skills,” says David Foote, chief analyst at Foote Partners. “In fact, they are all among the highest-paying skills for all 723 non-certified skills we report.” These skills bring in 16% to 22% of base salary, he says. AIOps, for example, saw an 11% increase in market value over the past year, now bringing in a premium of 20% over base salary, according to Foote data. MLOps now brings in a 22% premium. “Again, these AI skills have many uses of which the data center is only one,” Foote adds. The percentage increase in the specific subset of these skills in data centers jobs may vary. The Uptime Institute survey suggests that the higher pay is motivating workers to stay in the

ExtraHop looks to eliminate ‘extra hops’ in NDR stack

This deep visibility allows ExtraHop to provide insights across the entire network stack, from basic connectivity to application-level transactions. “The benefit of going all the way through Layer 7 is I can actually see a database transaction going through on the wire,” Vasani said. “If you have application teams complaining about database query latency, we can map it to what session was that tied to and what flows was it tied to from a network perspective and is this really an app server issue, or is it a network issue, or is it an endpoint issue?” The new sensor integrates with ExtraHop’s RevealX platform, feeding telemetry into the company’s cloud-scale ML/AI engine that powers its detection and analysis capabilities. “The sensor collects the telemetry, feeds it into an ML/AI engine that sits in the cloud, and then we layer in workflow engines on top to enable the various use cases,” Vasani said. In modern distributed enterprise environments, network visibility must extend beyond traditional data centers. ExtraHop’s all-in-one sensor is designed to address this reality with deployment options that span physical appliances, virtual machines and cloud environments. ExtraHop has both virtual and physical hardware appliances for sensor deployment. ExtraHop sensors can plug into a network through multiple methods including, Network Tap, SPAN (Switched Port Analyzer) port, packet broker or a cloud provider’s vTAP capabilities.

AI’s energy appetite drives interest in nuclear power

In its new report, Deloitte said that its analysis of figures from the World Nuclear Association, the American Nuclear Society, the U.S. Department of Energy, and others showed that new nuclear power could potentially meet about 10% of the projected increase in data center demand over the next decade, assuming capacity is also significantly expanded by between 35GW and 62GW, and 30% of the expansion is earmarked for data centers. “Nuclear energy presents a potential solution for meeting some of the growing electricity demands of data centers, with its reliable and clean energy profile,” Deloitte’s report said, noting five key advantages of the technology: Reliable baseload power: Nuclear reactors operate 24/7, regardless of the weather, providing the reliable power so important to data centers. In addition, Deloitte said, “Their capacity factor, exceeding 92.5%, outperforms other sources like natural gas (56%) and renewables like wind (35%) and solar (25%).” High energy density: A small amount of fuel generates a lot of power, which minimizes the need for fuel storage and transportation. “This efficiency can translate to a smaller physical footprint and enhanced sustainability,” Deloitte said. Scalable power output: A full-sized reactor typically generates 800 megawatts (MW) or more of electricity, which accommodates the needs of large data centers. Low carbon emissions: Nuclear power plants produce virtually no greenhouse gas emissions during operation. Enhanced land use efficiency: Compared to other energy sources, nuclear power plants require relatively little land. Gartner’s Johnson echoed these advantages, and also predicted that nuclear energy, and small modular reactors (SMRs) in particular, will “provide a viable answer” to the question of what to do when electricity demand exceeds supply. They can, he said, “ensure independence from grid power fluctuations by providing dedicated on-site power for large data centers.” However, both Gartner and Deloitte also highlighted challenges in

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.

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

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

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

Stay Ahead, Stay ONMINE