Lisa Morgan, Freelance WriterDecember 10, 202410 Min ReadMyrarte via Alamy StockAs businesses grow and tech stacks become more complex, scalability remains a top issue.Companies face significant challenges scaling across both physical and virtual spaces. While a holistic approach to operations across regions provides advantages, it also introduces complexity, says Dustin Johnson, CTO of advanced analytics software provider Seeq. The cloud can assist, but its not always a one-size-fits-all solution, especially regarding compute needs. Specialized resources like GPUs for AI workloads versus CPUs for standard processes are essential, and technologies like Kubernetes allow for effective clustering and scaling. However, applications must be designed to fully leverage these features, or they wont realize the benefits.The variety of technologies involved creates significant complexity.Today, a vertically integrated tech stack isnt practical, as companies rely on diverse applications, infrastructure, AI/ML tools and third-party systems, says Johnson. Integrating all these components -- ensuring compatibility, security, and scalability -- requires careful coordination across the entire tech landscape.A common mistake is treating scalability as a narrow technology issue rather than a foundational aspect of system design. Approaching it with a short-term, patchwork mentality limits long-term flexibility and can make it difficult to respond to growing demands.Related:Following are some more things that need to scale better in 2025.1. ProcessesA lot of organizations still have manual processes that prevent velocity and scale. For example, if a user needs to submit a ticket for a new server to implement a new project, someone must write the ticket, someone receives the ticket, someone must activate it, and then something must be done with it. Its an entire sequence of steps.Thats not a scalable way to run your environment so I think scaling processes by leveraging automation is a really important topic, says Hillery Hunter, CTO and GM of innovation at IBM and an IBM Fellow. There are a bunch of different answers to that [ranging] from automation to what people talk about, such as is IT ops or orchestration technologies. If you have a CIO who is trying to scale something and need to get permission separately from the chief information security officers, the chief risk officer or the chief data officer team, that serialization of approvals blocks speed and scalability.Organizations that want to achieve higher velocities should make it a joint responsibility among members of the C-suite.Related:You dont just want to automate inefficient things in your organization. You really want to transform the business process, says Hunter. When you bring together the owners of IT, information, and security at the same table, you remove that serialization of the decision process, and you remove the impulse to say no and create a collective impetus to say yes because everyone understands the transformation is mutual and a team goal.2. IT operationsIT is always under pressure to deliver faster without sacrificing quality, but the pressure to do more with less leaves IT leaders and their staff overwhelmed.Scalability needs to be done though greater efficiency and automation and use things like AIOps to oversee the environment and make sure that as you scale, you maintain your security and resiliency standards, says Hunter. I think re-envisioning the extent of automation within IT and application management is not done until those processes break. Its maybe not investing soon enough so they can scale soon enough.3. ArchitecturesIn the interest of getting to market quickly, startups might be tempted to build a new service from existing pre-made components that can be coupled together in ways that mostly fit but will demonstrate the business idea. This can lead to unintentionally complicated systems that are impossible to scale because of their sheer complexity. While this approach may work well in the beginning, getting business approval later to completely re-architect a working service that is showing signs of success may be very difficult.Related:First of all, be very careful in the architectural phase of a solution [because] complexity kills. This is not just a reliability or security argument, it is very much a scalability argument, says Jakob stergaard, CTO at cloud backup and recovery platform Keepit. A complex structure easily leads to situations where one cannot simply throw hardware at the problem this can lead to frustrations on both the business side and the engineering side.He advises: Start with a critical mindset, knowing that upfront investment in good architecture will pay for itself many times over.4. Data visibilityOrganizations are on a constant mission to monetize data. To do that they need to actively manage that data throughout the entire lifecycle at scale.While cloud computing has gained popularity over the past few decades, there is still a lot of confusion, resulting in challenges including understanding where your cloud data lives, what it contains, and how to ensure it is properly protected, says Arvind Nithrakashyap, co-founder and CTO at data security company Rubrik. When it comes to scalability one blind spot is unstructured and semi-structured data.Unstructured data poses a security risk, as it can contain sensitive business data or personally identifiable information. And since all unstructured data is shared with end-user applications using standard protocols over TCP/IP networks, its a prime target for threat actors. Since most companies have hybrid and multi-cloud implementations IT needs to understand where sensitive data is, where it is going and how it is being secured.One of the toughest hurdles for organizations whose unstructured data portfolio includesbillions of files, and/or petabytes of data, is maintaining an accurate, up-to-date count ofthose datasets and their usage patterns, says Nithrakashyap. [You need to understand] things [such as] how many files [exist], where they are, how old they are, and whether theyre still in active use. Without reliable, up-to-date visibility into the full spectrum of critical business files, your organization can easily be overwhelmed by the magnitude of your data footprint, not knowing where critical datasets are located, which datasets are still growing, [and] which datasets have aged out of use.5. SaaS service APIsAPIs are the glue that holds our modern software-driven world together. Keepits stergaard says his company sees bottlenecks on software-as-a-service APIs that vendors offer up for general use, from explicit throttling to slow responses, that are outright intermittent failures. For better and tighter integrations between systems, APIs need to scale to higher volume use.Fundamentally, an API that does not scale is pointless, says stergaard. For APIs to be useful we want them to be usable. Not a little bit, not just sometimes, but all the time and as much as we need. Otherwise, what's the point?Although it can be difficult to pinpoint a limiting factor, if user experience is any indication, it appears that some services are built on architectures that are difficult for the vendor to scale to higher volume use.This is a classical problem in computer science -- if a service is built, for example, around a central database, then adding more API front-end nodes may not do anything to improve the scalability of the APIs because the bottleneck may be in the central database, says stergaard. If the system is built with a central database being core to its functionality, then replacing that central component with something that is better distributed over many systems could require a complete re-write of the service from the ground up. In practical terms for real world services, making a service scale to higher volume use is often very different from just clicking the elastic scaling button on the cloud platform on which it runs.To scale a solution, it must be built on the simplest possible architecture, since architectural complexity is typically the main obstacle to scaling a solution. A complex architecture can make throwing hardware at a solution completely ineffective.6. Artificial intelligenceAs AI usage accelerates, cloud and cybersecurity scalability become even more critical.[M]ost companies are still in a discovery phase [with AI], and therefore what it takes to scale [in terms of] capabilities, cost, etc. is still not fully understood. It requires an approach of continuous learning and experimentation, with a strong focus on outcomes, to prioritize the right activities, saysOrla Daly, CIO at digital workforce transformation company Skillsoft.IT leaders must ensure alignment with business leaders on the desired outcomes and critical success factors. They also need to understand the skills and resources in the organization, define KPIs and fill key gaps.Teams who are not proactively managing the need for scale will find suboptimal decisions or runaway costs on one side, or [a] lack of progress because the enablers and path to scale are not defined, says Daly. Scaling technology is ultimately about enabling business outcomes, therefore continuing to tie activities to the company priorities is important. Its easy to get carried away by new and exciting capabilities, and innovation remains important, but when it comes to scaling, its more important to take a thoughtful and measured approach.7. Generative AIOrganizations are struggling with scaling GenAI cost-effectively. Most providers bill for their models based on tokens that are numerical representations of words or characters. The costs for input and output tokens differ. For example, Anthropics Claude 3.5 Sonnet charges $3.00 per million input tokens and $15 per million output tokens while OpenAIs gpt-4o model costs $2.50 per million input tokens and $10 per million output tokens. The two models are not equal and support different features, so the choice isnt as clear cut as which model is cheaper.GenAI model consumers must pick a balance between price, capability and performance. Everyone wants the highest quality tokens at the lowest possible price as quickly as possible, says Randall Hunt, CTO at leading cloud services company and AWS Premier Tier Services partner, Caylent.An additional charge exists around vectorization of data, such as converting images, text, or other information into a numerical format, called an embedding, that represents the semantic meaning of the underlying data rather than the specific content.Embedding models are typically cheaper than LLMs. [For instance,] Coheres Embed English embedding model is $0.10 per million tokens. Embeddings can be searched somewhat efficiently using techniques like [hierarchical navigable small world](HNSW) and cosine similarity, which isnt important, but it requires the use of database extensions or specialized datastores that are optimized for those kinds of searches -- further increasing cost. [A]ll of this cost is additive, and it can affect the unit economics of various AI projects.8. Operational technologydataCompanies are being flooded with data. This goes for most organizations, but its especially true for industrial companies that are constantly collecting operational technology (OT) data from equipment, sensors, machinery and more. Industrial companies are eager to integrate insights from OT and IT data to enable data-driven decision making based on a holistic view of the business.In 2025 and beyond, companies that can successfully give data context and make efficient and secure connections between diverse OT and IT data sources, will be best equipped to scale data throughout the organization for the best possible outcomes, says Heiko Claussen, chief technology officer at industrial software company AspenTech. Point-to-point data connections can be chaotic and complex, resulting in siloes and bottlenecks that could make data less effective for agile decision making, enterprise-scale digital transformation initiatives and AI applications.Without OT data fabric, an organization that has 100 data sources and 100 programs utilizing those sources would need to write and maintain 10,000 point-to-point connections. With an OT data fabric, that drops to 200 connections. In addition, many of these connections will be based on the same driver and thus much easier to maintain and secure.About the AuthorLisa MorganFreelance WriterLisa Morgan is a freelance writer who covers business and IT strategy and emergingtechnology for InformationWeek. She has contributed articles, reports, and other types of content to many technology, business, and mainstream publications and sites including tech pubs, The Washington Post and The Economist Intelligence Unit. Frequent areas of coverage include AI, analytics, cloud, cybersecurity, mobility, software development, and emerging cultural issues affecting the C-suite.See more from Lisa MorganNever Miss a Beat: Get a snapshot of the issues affecting the IT industry straight to your inbox.SIGN-UPYou May Also LikeWebinarsMore WebinarsReportsMore Reports

Todays IT departments have an amalgamation of DevOps, Waterfall, artificial intelligence, and OS/new release software, so quality assurance must be able to test and to verify the goodness of all these variegated systems. Yet, those of us who have led IT departments know that the QA function is habitually under-appreciated.Understanding that QA must broaden its reach to test such a broad spectrum of different systems, vendors have rolled out QA tools like the automated execution of test scripts that QA designs.This has generated a steady market in QA testing software, which Global Market Insights pinpointed at $51.8 billion in 2023, with a projected CAGR (compound annual growth rate) of 7% between 2024 and 2032.What IT departments should do now is strategize how a limited QA staff can best use these tools, while also developing the knowledge base and reach allowing them to cover the broad array of new applications and systems that QA is being asked to test.Performing QA With No Single Pane of GlassIf you are in system programming or network support, you know that there are over-arching software solutions that boast single pane of glass visibility. These systems provide an overall architecture that enables you to unify visibility of all of the different tools and functions that you have on a single screen. Not all IT departments invest in these expensive software architectures, but at least they do exist.Related:That isnt the case for quality assurance.In QA, the test bench is a hodgepodge of different tools and techniques spread out on a general tool bench. When a staffer performs QA, they pick whatever tools they choose to use from this tool bench based upon the type of application they are being called upon to test.If the application area to be tested is DevOps, QA is an iterative never done function that might use some test automation for workflow execution, but that also requires a high amount of collaboration between QA, development and end users until everyone arrives at a consensus that the application is production ready.In the AI environment, testing is also iterative and never finished. You work with development and user area subject matter experts to achieve the gold standard of 95% accuracy with what subject matter experts would conclude. Then you must periodically reaffirm accuracy because business conditions constantly change, and accuracy levels could fall.If the application is waterfall, it routes through the traditional path of development, unit test, integration test, regression test, deploy.Related:If the system is a new database or operating or infrastructure system release from a vendor, the new release is first simulated in a test environment, where it is tested and debugged. The new release gets installed into production when all testing issues in the simulated environment are resolved.Each of these test scenarios requires a different mental approach to QA and a different set of tools.Make QA a Strategic Function and Elevate its Standing?Test tool provider Hatica has stated,In the past, QA engineers were primarily focused on testing -- finding bugs and ensuring that the product worked as intended before it was released to users. However, this reactive approach to quality is no longer enough in todays environment. Before long, QA engineers will shift from being testers at the end of the process to quality strategists who are involved from the very beginning.In Agile and DevOps development, there already is an emerging trend for QA that confirms this. QA is immediately engaged in Agile and DevOps work teams, and the QA team provides as much input into the end-to-end DevOps/Agile process as development and end users. As IT departments move more work to Agile and DevOps, QAs role as a frontend strategist will expand.Related:However, in waterfall and new infrastructure release deployments, QAs role is more backend and traditional. It performs end of the line checkouts and is often not engaged in the initial stages of development. AI also presents a QA challenge, because a separate data science or subject matter expert group might do most of the system development and checkout, so QAs role is minimized.The Best Approach to QAThanks to the Agile/DevOps movement, QA now sees a more forward-thinking and strategic role. Yet at the same time, applications in the AI, waterfall, and infrastructure areas engage QA as more of a backend function.QA is also knee-capped by the lack of a single architecture for its tools, and by the brutal fact that most of the staff in QA departments are new hires or junior personnel. Quickly, these individuals apply for transfers into application development, database or systems, because they see these as the only viable options for advancing their IT careers.Understanding these realities, CIOs can do three things:1. Move QA into a more strategic position in all forms of application development. Like the IT help desk, QA has a long institutional memory of the common flaws in IT applications. If QA is engaged early in application development processes, it can raise awareness of these common flaws so they can be addressed up front in design.Accept as well that most QA staff members will want to move on to become a developer or an IT technical specialist and use QA as a grooming ground. To this end, the more QA gets engaged early in application planning and development, the more IT software knowledge QA staff will gain. This can prepare them for development or systems careers, if they choose to take these routes later.2. Ensure that QA staff is properly trained on QA tools. There is no uber architecture available for the broad assortment of tools that QA uses, so personalized training is key.3. Foster collaboration. In the Agile/DevOps environment, there is active collaboration between QA, development and end users. In AI development, CIOs can foster greater QA collaboration by teaming QA with IT business analysts, who often work side by side with user subject matter experts and data scientists. In new infrastructure release testing and in waterfall testing, more active collaboration should be fostered with system and application programmers.The more collaborative bridges you build, the more effectively your QA function will perform.

More than just fossils show us how humans have evolved through timeIvan M / Alamy Stock PhotoThis is an extract from Our Human Story, our newsletter about the revolution in archaeology.Sign up to receive it in your inbox every month.This month, Our Human Story turns 50 (months old). For the 50th instalment, I thought I would do something a little different: take stock of whats happened, and look ahead. I emailed 10 researchers, asking them two questions:What has been the biggest advance in human evolution of the past five years? This could

This is today's edition ofThe Download,our weekday newsletter that provides a daily dose of what's going on in the world of technology. We saw a demo of the new AI system powering Andurils vision for war James ODonnell One afternoon in late November, I visited a weapons test site in the foothills east of San Clemente, California operated by Anduril, a maker of AI-powered drones and missiles that recently announced a partnership with OpenAI. I went there to witness a new system its expanding today, which allows external parties to tap into its software and share data in order to speed up decision-making on the battlefield.If it works as planned over the course of a new three-year contract with the Pentagon, it could embed AI more deeply than ever before into the theater of war. Read the full story. How to use Sora, OpenAIs new video generating tool OpenAI has just released its video generation model Sora to the public. The announcement yesterday came on the fifth day of the companys shipmas event, a 12-day marathon of tech releases and demos. Heres what you should knowand how you can use the video model right now.James ODonnell This story is the latest in MIT Technology Reviews How To series, which helps you get things done. AIs hype and antitrust problem is coming under scrutinyThe AI sector is plagued by a lack of competition and a lot of deceitor at least thats one way to interpret the latest flurry of actions taken in Washington. The actionsfrom antitrust investigations to accusations of straight-up lyingrepresent an effort to hold the AI industrys hype to account in the final months before the Federal Trade Commissions chair, Lina Khan, is replaced when Donald Trump takes office.But while the FTC looks to have a far smoother transition of leadership ahead than most other federal agencies, at least some of Trumps frustrations with Big Tech could send antitrust efforts in a distinctly new direction. Read the full story.James ODonnell This story is from The Algorithm, our weekly newsletter giving you the inside track on all things happening in the fascinating field of AI. Sign up to receive it in your inbox every Monday. The must-reads Ive combed the internet to find you todays most fun/important/scary/fascinating stories about technology. 1 Google has built a powerful new quantum computing chip But it doesnt have any real-world applicationsyet. (Bloomberg $) + It takes five minutes to solve a problem that a traditional supercomputer could not master in 10 septillion years. (NYT $)+ Its a challenge the quantum field has been trying to crack for decades. (The Guardian)+We covered the work when it was a preprint in September.(MIT Technology Review) 2 Nvidia is being investigated by China It claims the chipmaking giant has violated anti-monopoly laws. (BBC)+ Nvidias biggest customer in the country? That would be ByteDance. (Insider $)+ Whats next in chips. (MIT Technology Review)3 TikTok has asked a US appeals court to halt the buy-or-sell law As it stands, the app faces a ban unless it finds a new owner by January 19. (TechCrunch)4 AI is still failing to deliver on its economic promises Is 2025 the year we finally start to see some results? (Quartz)+ The US AI industry is in desperate need of more sites with power grid access. (FT $)+ How to fine-tune AI for prosperity. (MIT Technology Review)5 The EUs competition rules are on the verge of a big shakeup A new boss means a new approach. (WSJ $)+ European regulators want to get to the bottom of a Meta and Google investigation. (FT $)6 Weight-loss drugs are making basic health truths obsolete A healthy diet and regular exercise is falling by the wayside. (The Atlantic $)+ Weight-loss injections have taken over the internet. But what does this mean for people IRL? (MIT Technology Review)7 This bionic leg is controlled by its wearers brain Prosthetic limbs are becoming much more capable. (New Yorker $)+ These prosthetics break the mold with third thumbs, spikes, and superhero skins. (MIT Technology Review)8 An AI can make a pretty decent Tokyo travel companionJust make sure you take its advice with a pinch of salt. (Wired $) + How to use AI to plan your next vacation. (MIT Technology Review)9 Reddit is testing a new AI search feature Which the sites users are unlikely to take kindly to. (Ars Technica) 10 Jeff Bezos has a dinner with Donald Trump in his diary Sounds cozy. (Insider $)Quote of the day Its like manna from heaven. Ari Morcos, chief executive of startup DatologyAI, explains to the Wall Street Journal why Reddits troves of text are so appealing to AI companies. The big story Inside the enigmatic minds of animals October 2022 More than ever, we feel a duty and desire to extend empathy to our nonhuman neighbors. In the last three years, more than 30 countries have formally recognized other animalsincluding gorillas, lobsters, crows, and octopusesas sentient beings. A trio of books from Ed Yong, Jackie Higgins, and Philip Ball detail creatures rich inner worlds and capture what has led to these developments: a booming field of experimental research challenging the long-standing view that animals are neither conscious nor cognitively complex. Read the full story. Matthew Ponsford We can still have nice things A place for comfort, fun and distraction to brighten up your day. (Got any ideas? Drop me a line or tweet 'em at me.) + It seems we have two types of laugh: one caused by tickling, and the other by everything else.+ 2024 was a strong year for fiction: check out some of the best new books.+ Theres something totally mesmerizing about this collection of old home videos.+ Ukrainian artist Oleg Dron specializes in expansive, haunting landscapes.

One afternoon in late November, I visited a weapons test site in the foothills east of San Clemente, California, operated by Anduril, a maker of AI-powered drones and missiles that recently announced a partnership with OpenAI. I went there to witness a new system its expanding today, which allows external parties to tap into its software and share data in order to speed up decision-making on the battlefield. If it works as planned over the course of a new three-year contract with the Pentagon, it could embed AI more deeply into the theater of war than ever before. Near the sites command center, which looked out over desert scrubs and sage, sat pieces of Andurils hardware suite that have helped the company earn its $14 billion valuation. There was Sentry, a security tower of cameras and sensors currently deployed at both US military bases and the US-Mexico border, and advanced radars. Multiple drones, including an eerily quiet model called Ghost, sat ready to be deployed. What I was there to watch, though, was a different kind of weapon, displayed on two large television screens positioned at the test sites command station. I was here to examine the pitch being made by Anduril, other companies in defense tech, and growing numbers of people within the Pentagon itself: A future great power conflictmilitary jargon for a global war involving competition between multiple countrieswill not be won by the entity with the most advanced drones or firepower, or even the cheapest firepower. It will be won by whoever can sort through and share information the fastest. And that will have to be done at the edge where threats arise, not necessarily at a command post in Washington. A desert drone test Youre going to need to really empower lower levels to make decisions, to understand whats going on, and to fight, Anduril CEO Brian Schimpf says. That is a different paradigm than today. Currently, information flows poorly among people on the battlefield and decision-makers higher up the chain. To show how the new tech will fix that, Anduril walked me through an exercise demonstrating how its system would take down an incoming drone threatening a base of the US military or its allies (the scenario at the center of Andurils new partnership with OpenAI). It began with a truck in the distance, driving toward the base. The AI-powered Sentry tower automatically recognized the object as a possible threat, highlighting it as a dot on one of the screens. Andurils software, called Lattice, sent a notification asking the human operator if he would like to send a Ghost drone to monitor. After a click of his mouse, the drone piloted itself autonomously toward the truck, as information on its location gathered by the Sentry was sent to the drone by the software. The truck disappeared behind some hills, so the Sentry tower camera that was initially trained on it lost contact. But the surveillance drone had already identified it, so its location stayed visible on the screen. We watched as someone in the truck got out and launched a drone, which Lattice again labeled as a threat. It asked the operator if hed like to send a second attack drone, which then piloted autonomously and locked onto the threatening drone. With one click, it could be instructed to fly into it fast enough to take it down. (We stopped short here, since Anduril isnt allowed to actually take down drones at this test site.) The entire operation could have been managed by one person with a mouse and computer. Anduril is building on these capabilities further by expanding Lattice Mesh, a software suite that allows other companies to tap into Andurils software and share data, the company announced today. More than 10 companies are now building their hardware into the systemeverything from autonomous submarines to self-driving trucksand Anduril has released a software development kit to help them do so. Military personnel operating hardware can then publish their own data to the network and subscribe to receive data feeds from other sensors in a secure environment. On December 3, the Pentagons Chief Digital and AI Office awarded a three-year contract to Anduril for Mesh. Andurils offering will also join forces with Maven, a program operated by the defense data giant Palantir that fuses information from different sources, like satellites and geolocation data. Its the project that led Google employees in 2018 to protest against working in warfare. Anduril and Palantir announced on December 6 that the military will be able to use the Maven and Lattice systems together. The militarys AI ambitions The aim is to make Andurils software indispensable to decision-makers. It also represents a massive expansion of how the military is currently using AI. You might think the US Department of Defense, advanced as it is, would already have this level of hardware connectivity. We have some semblance of it in our daily lives, where phones, smart TVs, laptops, and other devices can talk to each other and share information. But for the most part, the Pentagon is behind. Theres so much information in this battle space, particularly with the growth of drones, cameras, and other types of remote sensors, where folks are just sopping up tons of information, says Zak Kallenborn, a warfare analyst who works with the Center for Strategic and International Studies. Sorting through to find the most important information is a challenge. There might be something in there, but theres so much of it that we cant just set a human down and to deal with it, he says. Right now, humans also have to translate between systems made by different manufacturers. One soldier might have to manually rotate a camera to look around a base and see if theres a drone threat, and then manually send information about that drone to another soldier operating the weapon to take it down. Those instructions might be shared via a low-tech messenger appone on par with AOL Instant Messenger. That takes time. Its a problem the Pentagon is attempting to solve through its Joint All-Domain Command and Control plan, among other initiatives. For a long time, weve known that our military systems dont interoperate, says Chris Brose, former staff director of the Senate Armed Services Committee and principal advisor to Senator John McCain, who now works as Andurils chief strategy officer. Much of his work has been convincing Congress and the Pentagon that a software problem is just as worthy of a slice of the defense budget as jets and aircraft carriers. (Anduril spent nearly $1.6 million on lobbying last year, according to data from Open Secrets, and has numerous ties with the incoming Trump administration: Anduril founder Palmer Luckey has been a longtime donor and supporter of Trump, and JD Vance spearheaded an investment in Anduril in 2017 when he worked at venture capital firm Revolution.) Defense hardware also suffers from a connectivity problem. Tom Keane, a senior vice president in Andurils connected warfare division, walked me through a simple example from the civilian world. If you receive a text message while your phone is off, youll see the message when you turn the phone back on. Its preserved. But this functionality, which we dont even think about, Keane says, doesnt really exist in the design of many defense hardware systems. Data and communications can be easily lost in challenging military networks. Anduril says its system instead stores data locally. An AI data treasure trove The push to build more AI-connected hardware systems in the military could spark one of the largest data collection projects the Pentagon has ever undertaken, and companies like Anduril and Palantir have big plans. Exabytes of defense data, indispensable for AI training and inferencing, are currently evaporating, Anduril said on December 6, when it announced it would be working with Palantir to compile data collected in Lattice, including highly sensitive classified information, to train AI models. Training on a broader collection of data collected by all these sensors will also hugely boost the model-building efforts that Anduril is now doing in a partnership with OpenAI, announced on December 4. Earlier this year, Palantir also offered its AI tools to help the Pentagon reimagine how it categorizes and manages classified data. When Anduril founder Palmer Luckey told me in an interview in October that its not like theres some wealth of information on classified topics and understanding of weapons systems to train AI models on, he may have been foreshadowing what Anduril is now building. Even if some of this data from the military is already being collected, AI will suddenly make it much more useful. What is new is that the Defense Department now has the capability to use the data in new ways, Emelia Probasco, a senior fellow at the Center for Security and Emerging Technology at Georgetown University, wrote in an email. More data and ability to process it could support great accuracy and precision as well as faster information processing. The sum of these developments might be that AI models are brought more directly into military decision-making. That idea has brought scrutiny, as when Israel was found last year to have been using advanced AI models to process intelligence data and generate lists of targets. Human Rights Watch wrote in a report that the tools rely on faulty data and inexact approximations. I think we are already on a path to integrating AI, including generative AI, into the realm of decision-making, says Probasco, who authored a recent analysis of one such case. She examined a system built within the military in 2023 called Maven Smart System, which allows users to access sensor data from diverse sources [and] apply computer vision algorithms to help soldiers identify and choose military targets. Probasco said that building an AI system to control an entire decision pipeline, possibly without human intervention, isnt happening and that there are explicit US policies that would prevent it. A spokesperson for Anduril said that the purpose of Mesh is not to make decisions. The Mesh itself is not prescribing actions or making recommendations for battlefield decisions, the spokesperson said. Instead, the Mesh is surfacing time-sensitive informationinformation that operators will consider as they make those decisions.

On a podcast, Marc Andreessen said he has spent half his time at Mar-a-Lago since Election Day.He said he regularly discusses tech and economic policy with President-elect Donald Trump.Andreessen donated millions to the pro-Trump super PAC "Right for America" and other GOP candidates.Marc Andreessen said in a recent podcast appearance that since Election Day, he has spent roughly "half" his time at Mar-a-Lago discussing policy issues with President-elect Donald Trump.Speaking with Bari Weiss for an episode of her "Honesty" podcast, the venture capitalist and vocal Trump supporter said he's "not claiming to be in the middle of all the decision-making" but is trying to help shape policy in a second Trump administration."I've been trying to help in as many ways as I can," Andreessen said. "Trump brings out a lot of feelings in a lot of people. People have very strong views. And then there are many political topics that, you know, we're very deliberately not weighing it on."He added that he's "Not Mr. Foreign policy, or Mr. Abortion policy, or guns" because he's "not an expert on those things," but has instead focused his input on issues he does have experience in, like technology and economics."When I talk about these things, it's around tech policy, business, economics, and then, you know, the health of the country, the success of the country," Andreessen said.Representatives for Andreessen and Trump did not immediately respond to requests for comment from Business Insider.Andreessen was a vocal supporter of Trump's reelection bid. Business Insider previously reported he donated millions of dollars to a pro-Trump super PAC called "Right for America," and available data from OpenSecrets shows numerous donations to other Republican candidates and causes, such as the Republican Party of Michigan, Massachusetts, and Wisconsin, in the most recent election cycle.Last month, Andreessen and Ben Horowitz, the founders of the pair's powerful venture capital firm Andreessen Horowitz, celebrated Trump's win during an episode of their podcast, "The Ben and Marc Show.""It felt like a boot off the throat," Andreessen said of Trump's reelection. "Every morning I wake up happier than the day before."

UnitedHealthcare CEO Brian Thompson was fatally shot Wednesday outside a Midtown Manhattan Hilton.Luigi Mangione was arrested and faces a murder charge in the killing.The killing of the 50-year-old father of two sparked a manhunt in New York City and beyond.Luigi Mangione, the suspect in the shooting death of UnitedHealthcare CEO Brian Thompson, is fighting his extradition to New York City, where he'll face a murder charge in the Midtown Manhattan killing.The 26-year-old Ivy League graduate appeared for a hearing on Tuesday at Pennsylvania's Blair County Courthouse where his lawyer told the judge that Mangione was contesting his extradition.Mangione, who was arrested on Monday on local charges and later arraigned, made a bail request, which the judge denied during the hearing.The suspect will remain at Pennsylvania's Huntingdon State Correctional Institution as the extradition proceedings play out.Thomas Dickey, Mangione's lawyer, told reporters later on Tuesday that Mangione will plead not guilty to all the charges in Pennsylvania.In New York, an arrest warrant obtained by Business Insider shows Mangione faces a second-degree murder charge in connection to the fatal December 4 shooting of Thompson, a 50-year-old father of two from Minnesota. That charge carries a maximum sentence of life in prison.(A charge of first-degree murder is reserved for those accused of killing a law enforcement official or witness of a crime, or for when a murder is committed during the commission of another high-level crime, including robbery, rape, or kidnapping.) Mangione also faces four other charges related to the killing of the insurance CEO: two counts of criminal possession of a weapon in the second-degree, one count of second-degree criminal possession of a forged instrument, and one count of criminal possession of a weapon in the third-degree.During Mangione's Monday arrest in Altoona, Pennsylvania, officers found a three-page handwritten document "that speaks to both his motivation and mindset," New York Police Department Commissioner Jessica Tisch said at a press conference later that day.An internal NYPD report obtained by The New York Times gave the clearest view of the potential motive yet. Based on the so-called manifesto that was discovered, Mangione "likely views himself as a hero of sorts who has finally decided to act upon such injustices," the NYPD report said, as reported by the Times.Mangione "appeared to view the targeted killing of the company's highest-ranking representative as a symbolic takedown and a direct challenge to its alleged corruption and 'power games,' asserting in his note he is the 'first to face it with such brutal honesty,'" according to the NYPD report by the department's Intelligence and Counterterrorism Bureau, the Times reported.In a statement to BI, representatives for Nino Mangione a Maryland state legislator and a cousin of Mangione's declined to comment on the news of Mangione's arrest."Unfortunately, we cannot comment on news reports regarding Luigi Mangione," the statement read. "We only know what we have read in the media. Our family is shocked and devastated by Luigi's arrest."Recognized at a McDonald'sMangione was eating in an Altoona McDonald's when an employee recognized him from the several surveillance images that authorities released in the aftermath of Thompson's killing and called the police, New York police said at the Monday press conference.Altoona police found Mangione in the McDonald's with multiple fake IDs and a US passport, as well as a firearm and a suppressor "both consistent with the weapon used" in the shooting of Thompson in the heart of Manhattan, Tisch, the NYPD commissioner, said.The gun appeared to be a "ghost gun" that may have been made on a 3-D printer with the capability of firing a 9-millimeter round, NYPD Chief of Detectives Joseph Kenny said at the press conference.A Pennsylvania criminal complaint filed against Mangione said officers found a black 3-D-printed pistol and 3-D-printed silencer inside the suspect's backpack.When Altoona officers questioned Mangione and asked him if he had been to New York recently, he "became quiet and started to shake," the criminal complaint said.Clothing, including a mask, was also recovered "consistent with those worn" by the suspect wanted for Thompson's killing, along with a fake New Jersey ID matching the ID that the murder suspect used to check into a Manhattan hostel before the attack, Tisch said.Based on the handwritten document that police found on Mangione, according to Kenny, "it does seem that he has some ill will toward corporate America."During an interview Tuesday morning on NBC's "Today" show, Tisch said the "manifesto" revealed "anti-corporatist sentiment" and "a lot of issues with the health care industry.""But as to like particular, specific motive that'll come out as this investigation continues to unfold over the next weeks and month," the NYPD commissioner said.NBC News and The New York Times, each citing an unnamed senior law enforcement official, reported that the handwritten document read in part: "These parasites had it coming.""I do apologize for any strife and trauma, but it had to be done," it added, according to the reports.Police believe that Mangione acted alone.NYPD investigators traveled to Altoona on Monday to interview Mangione after Altoona officers took him into custody.Blair County District Attorney Peter Weeks said at Mangione's Pennsylvania arraignment that Mangione was carrying $10,000 in cash, including foreign currency, according to the Associated Press.Mangione disputed the amount in court. NYPD released images of the person of interest in Brian Thompson's killing. DCPI/NYPD Mangione was active on social mediaOn the social media site X, Mangione posted and amplified posts about technological advances like artificial intelligence. He also posted about fitness and healthy living.He frequently retweeted posts by the writer Tim Urban and commentator Jonathan Haidt about the promise and perils of technology.He also appeared to be a fan of Michael Pollan, known for his writing about food and ethics, and lab-grown meat. At the top of his profile was a header image with three images: a photo of himself, smiling, shirtless on a mountain ridge, a Pokemon, and an x-ray with four pins or screws visible in the lower back.Mangione founded a company called AppRoar Studios in 2015 while still in high school. AppRoar released an iPhone game called Pivot Plane that is no longer available.The two other cofounders of AppRoar could not be reached for comment.Mangione's X account has been deactivated. A spokesperson for YouTube said his three accounts on the platform were also terminated, but that they had not been active for about seven months.A manhuntMangione's arrest follows a nearly week-long manhunt.According to police, Manigone was born and raised in Maryland, and has ties to San Francisco, California. His last known address was in Honolulu, Hawaii.The New York Post, citing law-enforcement sources, reported that Mangione's mother reported him missing in mid-November.Kenny, the NYPD's chief of detectives, said that Manigone has no prior arrest history in New York and no known arrests in the US.A Luigi Mangione with a matching birthday and address received a citation for simple trespass for entering a forbidden area of a state park in Hawaii in November 2023. He pleaded no contest and paid a $100 fine."For just over five days, our NYPD investigators combed through thousands of hours of video, followed up on hundreds of tips, and processed every bit of forensic evidence DNA, fingerprints, IP addresses and so much to tighten the net," Tisch said at Monday's press conference announcing the arrest of Manigone.Thompson was shot multiple times on a Midtown sidewalk as he was walking toward the Hilton hotel. He was steps away from a side entrance to the hotel where he was set to speak at UnitedHealth Group's investor conference when a hooded gunman opened fire on him from behind.The chief executive of the nation's largest health insurer was struck at least once in the back and at least once in the right calf, police said.Surveillance footage showed the gunman firing his weapon as Thompson, wearing a blue suit jacket, walked several feet in front of him.The gunman fled the scene, first on foot and then on an electric bike, which he rode into Central Park before ultimately escaping from New York City, police said.Shell casings and bullets found at the scene had the words "deny," "defend," and "depose" written on them, according to multiple reports citing unnamed sources. BI couldn't independently confirm these details.In the aftermath of the attack, the NYPD offered a $10,000 reward for tips leading to the gunman's arrest, with the FBI offering a reward of up to $50,000 for information leading to his arrest and conviction.A spokesperson for UnitedHealth Group, the parent company of UnitedHealthcare, reacted to news of Manigone's arrest in a statement to BI, saying: "Our hope is that today's apprehension brings some relief to Brian's family, friends, colleagues and the many others affected by this unspeakable tragedy. We thank law enforcement and will continue to work with them on this investigation. We ask that everyone respect the family's privacy as they mourn."This story is developing and will be updated.

On Monday, police in Altoona, Pennsylvania, arrested Luigi Mangione in connection with last weeks shooting death of UnitedHealthcare CEO Brian Thompson. When Mangiones identity became public, the online reaction around Thompsons death went into overdrive; unlike most shootings, this one has brought a stream of support for the suspected killer rather than for the victim, and Mangiones capture has only intensified that sentiment. Related:Mangione, 26, has since been charged with second-degree murder in New York, among other crimes, and has been valorized as a folk hero in many corners of the internet, though the killing of Thompson, 50, has also been widely condemned. After Mangine was arrested and identified, his following on X went from a few dozen followers to more than 300,000 overnight. From a flood of supportive memes on Instagram and X to a shooter lookalike contest this weekend in Manhattans Washington Square Park, making jokes about Thompsons death somehow came to be considered acceptable and appropriate.The widespread interest in Mangione demonstrates just how much built-up hate there is toward insurance companies: Americans are infuriated about the costs and complications of health care coverage, and Thompsons murder has brought that anger to the surface. To try to better grasp the reaction among Americans, Today, Explained reached out to journalist Bob Herman, who covers the business of health care at Stat News and has co-reported several investigations about UnitedHealthcare.Today, Explained co-host Sean Rameswaram spoke with Herman about UnitedHealthcares business practices, their place in the health care market, and why they have been a target of so much anger. An excerpt of his conversation with Herman, edited for length and clarity, is below. In the full podcast we also get into who Mangione is and what we know about his motives, why it took police five days to find him, and how they eventually did, so listen to Today, Explained on Apple Podcasts, Spotify, or wherever you get podcasts.Bob HermanUnitedHealthcare is the largest health insurance company in America, and it is part of a bigger company called UnitedHealth Group. UnitedHealthcare covers older adults on Medicare Advantage, they cover the poor who are on Medicaid, they sell Obamacare plans, and theyre also the plans that employers offer to their workers. Weve had a whole series this year called Health Cares Colossus that has looked at this massive company and how it has its tentacles everywhere. A lot of people recognize it for UnitedHealthcare, the insurance company. It is not just a health insurance company thats what its most known for, thats why its often reviled but people dont recognize all the other components that it owns. It owns a ton of medical practices. It either employs or is affiliated with one out of every 10 doctors in the country. It is the largest claims processor. They are everywhere.Sean RameswaramWhat was your reaction when you saw that Brian Thompson had been shot last week? Bob HermanIve never covered a homicide on my beat. Normally, this is a beat filled with white-collar crime, so this definitely has been outside of the norm. My colleague and I were watching UnitedHealth Groups Investor Day on Wednesday morning, just kind of a routine thing. Its a big meeting every year all kinds of investors and shareholders get together and they talk about their earnings for the year and its a big cheerleading routine. Normally Brian Thompson appears pretty early on, as the head of the insurance division for United. And then about an hour in, thats when I first got a text of a New York Post story saying that Thompson had been shot and killed. So a little bit more than an hour into it, thats when Andrew Witty, the CEO of UnitedHealth Group, the parent company, came back out and canceled the rest of the conference. We had never seen anything like it.Sean RameswaramAlmost immediately, people were celebrating this homicide. What was your reaction to that?Bob HermanThe reactions were pretty morbid. Pretty grim. The dancing on the death of somebody was pretty vile. But anyone who covers health care knows that people are fed up with the system. This is not new. This has been going on for decades. Even when the Affordable Care Act, or Obamacare, came into play 15 years ago, people still hated the system and its persisted since then. So peoples frustrations have been bubbling under the surface for so long. To say you were surprised by all the reactions, then I think maybe you had your head buried in the sand a little bit. Sean RameswaramBefore we talk about what this company does that might upset people, can you just tell us about the company generally? Bob HermanUnitedHealthcare is the largest health insurance company in America. They cover more people than anyone [its coverage] includes Medicare Advantage, Medicaid, Obamacare, and all the different types of employer plans out there. And theres a lot. Sean RameswaramWhich I imagine makes them pretty powerful in this market.Bob HermanYeah. Now, a lot of health care dynamics are local. The fights that often occur between health insurers and providers, like hospitals and doctors, are all about who has market share in a specific area. United might not always have the biggest market share in a given area, but nationally they are very powerful, they are very profitable. Theres no way of getting around them in pretty much any market. They exist almost everywhere in some capacity. Sean RameswaramAnd what do they do with that power? How do they throw it around? Bob HermanIf you have a UnitedHealthcare plan, you want them to negotiate better rates for you. Thats their primary function. And so they use some of that to try and drive better deals with hospitals and doctors and other types of facilities. Theyre also an insurance company and they happen to deny or delay claims as well. As an insurance company, the easiest way to make money is to make sure youre paying out fewer dollars. Now, there are laws in place, especially a federal law that says you have to spend a certain amount of your premium dollars on care for people. You just have to do it. But insurers have an incentive to stay as close to those thresholds as possible. They dont want to have to pay out more than they absolutely have to.Sean RameswaramYou just, of course, used two of the operative words in this story because as we found out, the shooter had inscribed three words on his bullets: deny, delay, and depose. Is this company known for its denials? Bob HermanEvery health insurance company out there is known for their denials, to some degree. Specific rates of denials are tough to come by, but insurers especially the for-profit ones that [trade] on the stock market have an incentive to deny care. There have been so many issues with denials over the years. My colleague Casey Ross and I reported on a big one within Medicare Advantage, the program for older adults. UnitedHealthcare is the biggest [provider] in Medicare Advantage, and over the past several years, theyve been using algorithmic prediction, predictive tools, and artificial intelligence to basically ramp up denials specifically in post-acute care, the care that somebody gets when they leave the hospital. So especially on that side of the coin, post-acute care denials have been a very big issue for United. Sean RameswaramWhat exactly does that mean, that theyre using algorithms and AI to deny Medicare Advantage [claims]? How do they do that? Bob HermanSo lets say someone goes to the hospital and then the hospital says, okay, you know, youre ready for physical therapy, occupational therapy, speech therapy. Lets send you to a rehab facility or a nursing home. So a person will go there and theyll start their physical therapy, and behind the scenes, UnitedHealthcare has used a tool called NaviHealth. Theres an algorithm within the company that looks at the patients demographics how sick they are, their history and tries to come up with some kind of prediction of how much time theyll need in that nursing home. Lets say its 16 days. Thats what the algorithm says after 16 days, you should be good. Now, if its used as a guide, thats fine. But in many cases, we found documents that said that United told their case managers, You have to stick to the algorithm.And thats where it becomes a problem, because if youre saying this algorithm spits out 16 days for somebody and theyre not ready to go home on the 16th day, if they cant even go to the bathroom themselves, if they still cant walk around but the algorithm says its time to ship them out, thats where its a problem. And thats whats happened. Then families are left with the decision of, do I pay out of pocket to stay at this nursing home to get the care that my mom or dad or grandparent needs? Or do I take them home with me and then risk having them fall or get hurt again and have to go back to the hospital or worse? So thats how the algorithms play in those types of situations. Sean RameswaramAnd people know this is happening and theyre mad about it? Bob HermanOn the Medicare Advantage side, most people actually dont. Theyre not aware as much about this because the algorithm happens behind the scenes. Its not like families are getting a sheet of paper saying, Hey, our algorithm says you have this much time here. They really dont find out about this until the nursing home says, Hey, your insurance is up and we have to kick you out now. The denials are very front and center in other insurance plans where they say, I know I need this back surgery or I need this prescription, and then United will come along with maybe a prior authorization which says, Hey, doctor, fill out more paperwork to make sure that this person needs this procedure or needs this drug. And then maybe theyll come back with a denial that says, Were not going to cover this procedure or this prescription drug. Thats where its most front and center for people, thats where a lot of the outrage comes from: those widespread delays and denials. Sean RameswaramWere talking about our eldest citizens who are often on the receiving end of these algorithms or AI initiatives that tell them theyve run out of care. Bob HermanAt least in our reporting it is for the Medicare Advantage it is the oldest, its the frail, often the poor, the oldest people in this country who often have no idea that this is going on. Once they find out, they can appeal; anyone can appeal any denial. But it is such an arduous process. If youre sick or injured, thats not something you want to be doing, and you might not have family to help you out either. So it is very clearly a problem. Sean RameswaramIs this a UnitedHealthcare-specific problem or is this a systemic problem? It feels like a systemic problem. Bob HermanThis is systemic. This is not isolated to just UnitedHealthcare. UnitedHealthcare gets the most criticism and heat for this because they are the largest and theyre a very common provider for any workplace plan. But there are other large insurers: Cigna, Aetna, all the Blue Cross Blue Shield plans, Humana. This is just how US health insurance works. This is a systemic issue, especially for the insurance companies that are on the stock market. They have a duty to make money for shareholders. And one of the ways that they do that is by making sure that they pay out fewer claims. The most-watched number on every earnings call for an insurer is whats called the medical loss ratio. Thats a number that says how much money from our premiums we spend on medical care, and lower is better. If its higher than expected, Wall Street freaks out. I think that kind of tells you a lot. Sean RameswaramThis has been an insane week for this industry. But do you think anything changes now, other than executives are going to have more security? Bob HermanCould this be a galvanizing event to broader health care reform? Its certainly possible, because the American public has made their voices very clear here. But this is completely dependent on a new administration, a new Congress. This is a federal policy issue. And if Congress doesnt act, then youre just going to continue to see more of this unless companies start to make changes on their own. But if they do, itll be around the edges. Itll be tinkering.Youve read 1 article in the last monthHere at Vox, we're unwavering in our commitment to covering the issues that matter most to you threats to democracy, immigration, reproductive rights, the environment, and the rising polarization across this country.Our mission is to provide clear, accessible journalism that empowers you to stay informed and engaged in shaping our world. By becoming a Vox Member, you directly strengthen our ability to deliver in-depth, independent reporting that drives meaningful change.We rely on readers like you join us.Swati SharmaVox Editor-in-ChiefSee More:

President-elect Donald Trump is reportedly considering deporting some immigrants to countries other than their own. If he tries to do so, it wont be the first time. Like before, however, he would probably face legal challenges. According to NBC News, Trump is considering sending immigrants whose home countries will not accept US deportees to third countries including Turks and Caicos, the Bahamas, Panama, and Grenada. Currently, many immigrants from so-called recalcitrant countries are simply released into the US since there is nowhere to send them. Its not immediately clear what legal mechanism Trump intends to rely on to carry out these deportations to third countries. A representative for the Trump transition team did not respond to a request for comment. A regulation and a law currently give the executive branch some ability to deport immigrants to third countries; however, the legality of both is an open question. During his first term, Trump previously sought to use executive power to send asylum seekers of various nationalities to Guatemala under what he called an Asylum Cooperative Agreement. Under the agreement, migrants who passed through Guatemala before arriving in the US were sent back if they did not first seek protection there. The ACLU filed a lawsuit challenging the policy, but that suit was never resolved: The government stopped enforcing the policy during the pandemic and President Joe Biden was elected. The rule remains on the books, however. If Biden does not rescind it before he leaves office, the incoming Trump administration could use it to deport people to the countries under consideration if it survives in court, and if the US can broker similar agreements with those countries. Alternatively, Trump could try to invoke federal immigration law allowing the removal of immigrants to third countries in certain circumstances, such as when they cannot be returned to their country of origin and the third country is deemed to be safe for them. The ACLU has challenged Bidens use of this law to fast-track deportations of Venezuelans to Mexico without their consent. The outcome of the lawsuit may determine the kind of powers Trump may have to carry out his plans.Either way, the Trump administration would have to ensure that immigrants would be sent to a country where they will be safe, as is required under US and international law. Folks are supposed to be safe from persecution and torture and [any] procedure has to include adequate screening for fear of return and a fair process, Katrina Eiland, deputy director of the ACLUs Immigrants Rights Project, said. To the extent [the Trump administration is] incentivized to take shortcuts, thats a huge problem and something that the ACLU and other allies Im sure would be prime to sue over. Recalcitrant countries, briefly explainedThe US had designated 13 countries as recalcitrant as of 2020, including Russia, China, Cuba, and Iran. It has not publicly updated that list in the years since, and immigrants have been arriving in the US in increasing numbers from some of those countries. For instance, apprehensions of Chinese nationals at the US southern border jumped from less than 2,000 in fiscal year 2022 to over 36,000 in 2024. Many of them are fleeing economic hardship and political oppression following the countrys strict pandemic-era lockdowns. But China has been reluctant to accept its own citizens: The US sent a large deportation flight to China in July for the first time in six years. Though Venezuela was not previously on the list of recalcitrant countries, it also stopped accepting deportation flights from the US in February following the implementation of American sanctions. While the US was previously only returning a fraction of the millions fleeing Venezuelas dictatorship, the Biden administration saw the deportation flights as a deterrent to further migration. US immigration agents recorded more than 300,000 encounters with Venezuelans in fiscal year 2024.All of these people, who number in at least the hundreds of thousands, could be targets for a deportation program that sends immigrants to third countries under Trump.Can Trump actually deport people to third countries?Whether Trump can deport people to third countries in large numbers may depend on what happens in the ACLUs pending lawsuits. But, again, existing legal authorities could allow him to carry out at least some of these removals.The ACLU has argued in its lawsuit challenging the rule underlying the USs agreement with Guatemala that the agreement does not provide for sufficient screening to determine whether an immigrant would face credible fear of persecution in Guatemala. Under US and international law, immigrants cannot be returned to places where they would face such credible fear. At the time the agreement was established, Guatemala had the ninth-highest homicide rate worldwide, at about 26 deaths per 100,000 inhabitants, and the State Department had issued a travel warning for US citizens in Guatemala.The rule also claimed that asylum seekers would only be sent to countries where they have access to a full and fair procedure for determining a claim to asylum or equivalent temporary protection. The Trump administration certified that Guatemalas legal framework met that standard despite what Eiland called a total dearth of evidence in the administrative record, and in fact a lot of evidence to the contrary.Its not clear whether the same legal arguments would apply to any similar agreements Trump might broker with Turks and Caicos, the Bahamas, Panama, and Grenada. But in the meantime, Biden still has an opportunity to rescind the underlying rule that would allow additional agreements to be implemented.With a little over a month to go till inauguration, they may do nothing, in which case the rule is still there. Theoretically, the Trump administration could come in and sign new agreements, Eiland said. In its other lawsuit challenging the deportations of Venezuelans to Mexico, the ACLU has argued that such a use of the third-country removal authority is unprecedented and will result in removal to situations in which noncitizens are likely to face persecution or torture. The law lays out a detailed process for determining when an immigrant can be sent to a third country, and the ACLU has argued that the Biden administration is not abiding by it.If the courts uphold Bidens use of the law, that would potentially open the door for Trump to do the same for additional citizens of recalcitrant countries, giving him another tool through which he could carry out his plans for mass deportations. Youve read 1 article in the last monthHere at Vox, we're unwavering in our commitment to covering the issues that matter most to you threats to democracy, immigration, reproductive rights, the environment, and the rising polarization across this country.Our mission is to provide clear, accessible journalism that empowers you to stay informed and engaged in shaping our world. By becoming a Vox Member, you directly strengthen our ability to deliver in-depth, independent reporting that drives meaningful change.We rely on readers like you join us.Swati SharmaVox Editor-in-ChiefSee More:

The history of video games is, to an extent, a history of subtle iterations of other peoples ideas. The interstellar success of Taitos Space Invaders spawned the entire shoot-em-up genre, with titles such as Galaxian, Phoenix and Gorf taking the basic idea and adding new features. Later, 1984s Karate Champ begat the fighting game craze and Tetris brought us falling-block puzzlers. Its the way things have always worked: adapt, expand, hand on the baton. Its just that theres a subtle yet deep gulf between imitation and inspiration and not all titles manage to cross it.Marvel Rivals, the latest live service game from Chinese mega publisher NetEase, is Overwatch with Marvel characters. Thats not just the elevator pitch, thats exactly what it is. A colourful roster of comic book characters with differing skills meets in a series of sci-fi arenas to do team-based battle over a small selection of play modes. Machine-gun-toting vanilla guy Punisher is Overwatchs Soldier 76 with a hint of Bastion; God-like healer Adam Warlock is the male Mercy; and as a fist-thudding tank, Hulk is just rampaging gorilla Winston with less hair. Gaming site GamesRadar has even provided a handy guide to show players which of the Marvel cast members are most analogous to their Overwatch faves.Marvel Rivals. Photograph: Games PressMany of the well-worn tropes and abilities of the genre have at least been remixed to fit the Marvel universe and playing as these familiar legends adds undeniable appeal. Whether it is bludgeoning enemies with Thors hammer, catapulting exploding acorns as Squirrel Girl or lobbing Captain Americas shield into Black Panthers body armour, Rivals captures the comic book dynamism of this famed cast to such an extent that major skirmishes look like the most exciting scenes from the X-Men 97 cartoon. Its good too that all 33 heroes are available for free from the beginning. Theres a store and a battle pass, naturally, but these will currently only get you alternative outfits, emotes and other accessories; and completing daily missions and seasonal story objectives earns currency to buy this kind of stuff without paying a penny.Furthermore, the game does have one big new feature Team-Ups which unlock extra hero abilities when at least two players on the same side choose complementary characters. Theres the Symbiote Bond between Venom, Spider-Man and Peni Parker, which allows the latter two to channel the formers alien powers, and theres Ragnarok: Rebirth, which lets Hela heal or revive Thor or Loki. The kinships can be be a real boost to tactical play.Marvel Rivals. Photograph: NetEase GamesBut in many ways, Rivals reflects that key tenet of the hero shooter design bible: for every plus there must be a minus. With its huge slate of Marvel super freaks and their Team-Up powers, the game feels extremely unbalanced at times. It is hard to counter characters such as Storm and Iron Man when they can stay in the air for a whole match, picking off enemies from afar and avoiding most of the incoming fire. Big-hitters such as Venom and Moon Knight tend to completely dominate any area theyre fighting in, often at the expense of melee-based combatants who need to get in close to deal major damage. I never thought Wolverine would turn out to be one of the more subtle and refined characters in a Marvel cast, but here we are.The game is undeniably lavish to look at and interact with. The user-interface design around menu systems and information screens is exceptional; the destructible locations sparkle with detail; and the characters are beautifully re-created. Once again, however, theres a downside. During the chaos of a superhero riot, with explosions and magic attacks and hilarious quips firing off simultaneously, its tough to tell what youre damaging and, conversely, whats damaging you until its far too late.Buffs and nerfs will surely come for these characters in time, evening out the balance, and players will start learning how to combine team members more strategically. But even if the balancing issues are addressed, what were left with is the video game equivalent of a folkloric changeling, a supernaturally accurate substitute designed to entrap those who loved the original. The question is: can we really blame Rivals for walking so close to Overwatch that it could conceivably be served with a restraining order? As failed hero shooters Hyenas, Concord and xDefiant have all recently shown, the brutal economics of the live service market demand absolute fealty to established norms. Tagging on a massive global licence doesnt hurt either.Rivals, like many other highly polished, super-focused franchise expansions, is entertaining, gorgeous and well made. But its existence bodes ill for the mainstream games industry and those who work in it. It says that to be successful, especially in the live service sector (where so much investment is going), broadening or challenging a genre isnt necessary. You simply need to replicate and re-franchise, tossing a few low denomination coins at the concept of innovation. Meanwhile, studios that strike out with new ideas or original characters are doomed to failure: millions of dollars lost, jobs gone, game over.Rivals is crammed with Stan Lee superheroes, but its message about the total and utter Funko-Pop-ification of games is as bleak as a Charles Burns graphic novel.Marvel Rivals is out now; free to play, with in-game purchases