GamesIndustry.biz presents The Year In Numbers 2024Our annual infographic reveals the global games market value, best-selling games, biggest mobile markets, and more Feature by James Batchelor Editor-in-chief Published on Dec. 20, 2024 Tis the season to be jolly pouring over pie charts, bar graphs, tables and any other statistics we can throw at you.Yes, the approach of the holiday break means the annual GamesIndustry.biz Year In Numbers infographic is ready to go, perfectly capturing the current state of the video games market in one (admittedly very long) JPG image.As always, a huge thank you to the experts that share their data and help us compile this together. In the infographic below, you will find:The global games market value (courtesy of Newzoo)Total boxed vs digital revenues (also via Newzoo)The biggest mobile games, publishers and countries, both by revenue and downloads (thank you, Sensor Tower)The best-selling games in the UK (from Nielsen GfK), US (from Circana) and Japan (from Famitsu)The most covered games and companies, in terms of number of articles (provided by ICO via Footprints, based on data from over 18,000 websites)The biggest influencers on YouTube, most watched trailers and game promo videos, and the most popular video games on TikTok (courtesy of Fancensus)With that, GamesIndustry.biz presents The Year In Numbers 2024:

Natural Language processing uses large language models (LLMs) to enable applications such as language translation, sentiment analysis, speech recognition, and text summarization. These models depend on human feedback-based supervised data, but relying on unsupervised data becomes necessary as they surpass human capabilities. However, the issue of alignment arises as the models get more complex and nuanced. Researchers at Carnegie Mellon University, Peking University, MIT-IBM Watson AI Lab, University of Cambridge, Max Planck Institute for Intelligent Systems, and UMass Amherst have developed the Easy-to-Hard Generalization (E2H) methodology that tackles the problem of alignment in complex tasks without relying on human feedback.Traditional alignment techniques rely heavily on supervised fine-tuning and Reinforcement Learning from Human Feedback (RLHF). This reliance on human capabilities serves as a hindrance when scaling these systems, as collecting high-quality human feedback is labor-intensive and costly. Furthermore, the generalization of these models to scenarios beyond learned behaviors is challenging. Therefore, there is an urgent need for a methodology that can accomplish complex tasks without requiring exhaustive human supervision.The proposed solution, Easy-to-Hard Generalization, employs a three-step methodology to achieve scalable task generalization:Process-Supervised Reward Models (PRMs): The models are trained on simple human-level tasks. These trained models then evaluate and guide the problem-solving capability of AI on higher-level complex tasks.Easy-to-Hard Generalization: The models are gradually exposed to more complex tasks as they train. Predictions and evaluations from the easier tasks are used to guide learning on harder ones.Iterative Refinement: The models are adjusted based on the feedback provided by the PRMs.This learning process with iterative refinement enables AI to shift from human-feedback-dependent models to reduced human annotations. Generalization of tasks that deviate from the learned behavior is smoother. Thus, this method optimizes AIs performance in situations where human engagement becomes obscure.Performance comparison shows significant improvements on the MATH500 benchmark, a 7b process-supervised RL model achieved 34.0% accuracy, while a 34b model reached 52.5% accuracy, using only human supervision on easy problems. The method demonstrated effectiveness on the APPS coding benchmark as well. These results suggest comparable or superior alignment outcomes to RLHF while significantly reducing the need for human-labeled data on complex tasks.This research addresses the critical challenge of AI alignment beyond human supervision by introducing an innovative, easy-to-hard generalization framework. The proposed method demonstrates promising results in enabling AI systems to tackle increasingly complex tasks while aligning with human values. Notable strengths include its novel approach to scalable alignment, effectiveness across domains such as mathematics and coding, and potential to address limitations of current alignment methods. However, further validation in diverse, real-world scenarios is necessary. Overall, this work marks a significant step toward developing AI systems that can safely and effectively operate without direct human supervision, paving the way for more advanced and aligned AI technologies.Check out the Paper and GitHub Page. All credit for this research goes to the researchers of this project. Also,dont forget to follow us onTwitter and join ourTelegram Channel andLinkedIn Group. Dont Forget to join our60k+ ML SubReddit. Trending: LG AI Research Releases EXAONE 3.5: Three Open-Source Bilingual Frontier AI-level Models Delivering Unmatched Instruction Following and Long Context Understanding for Global Leadership in Generative AI Excellence.The post How AI Models Learn to Solve Problems That Humans Cant appeared first on MarkTechPost.

Pokmon Trading Card Game Pocket has released its third player-versus-player event today, December 20, which awards Emblems based on the Mythical Island expansion.The Mythical Island Emblem Event follows the same format as the first PvP event released for the digital trading card game, simply challenging players with racking up 45 total wins over three weeks to obtain the best rewards. The event ends on January 10.Players can obtain Pack Hourglasses just for competing in the event, however. Three are rewarded for participating in one battle and three battles, six are rewarded for participating in five, and 12 are rewarded for participating in 10.Every Alternate Art 'Secret' Card in Pokmon TCG Pocket: Mythical IslandIGN's Twenty Questions - Guess the game!IGN's Twenty Questions - Guess the game!To start:...try asking a question that can be answered with a "Yes" or "No".000/250Shinedust is on offer for those who win battles: one win rewards 50, three rewards 100, five rewards 200, 10 rewards 500, 25 rewards 1,000, and 50 rewards 2,000. Players will also obtain Emblems for winning, earning a white emblem for one win, a bronze emblem for five wins, a silver emblem for 25 wins, and a gold emblem for 45 wins.Mythical Island arrived as the first major expansion for Pokmon TCG Pocket and with it came a meta shake up. Pikachu ex appears to have fallen off a little thanks to a number of cards being added which appear designed specifically to take it down, while Mewtwo ex received a subtle boost.Celebi ex has also emerged as a new threat to the meta, but now the initially overlooked Mew ex is also entering the spotlight.Pokmon TCG Pocket arrived October 30 and is a certified hit for developer Creatures Inc. and The Pokmon Company, having earned an estimated $200 million in its first month across more than 60 million downloads.This huge amount of money comes as Pokmon TCG Pocket follows the standard mobile and free to play game model, flooding players with rewards in the first few days before soon drying up, with spending real world money the only real way to re-experience that early thrill outside of the occasional set drop like Mythical Island.Completing Genetic Apex, the first set of cards which totals 226 officially but also contains 60 rare alternate art cards, will take players not spending money around two years according to one estimate, while those looking to make it rain can wrap up the collection after dropping around $1,500.Ryan Dinsdale is an IGN freelance reporter. He'll talk about The Witcher all day.

28 Years Later, the long-awaited sequel to cult classic horror zombie movie 28 Days Later, doesnt come out until June 2025, but we now have the announcement of the second film in the planned trilogy alongside its release date.28 Years Later: The Bone Temple from Columbia Pictures launches January 16, 2026. Its directed by Nia DaCosta (The Marvels, Candyman), with a screenplay from original 28 Days Later writer Alex Garland. The producers are Danny Boyle, who directed 28 Days Later and is directing 28 Years Later, Alex Garland, Andrew Macdonald, Peter Rice, and Bernard Bellew. Oppenheimer star Cillian Murphy, the protagonist of 28 Days Later, is executive producing.The Bone Temple, then, releases just half a year after 28 Years Later as the second film in the trilogy. The third film has yet to be announced.28 Days Later was one of Cillian Murphys breakout roles back in 2002. He played Jim, a bicycle courier who woke from a coma to find himself in a post-apocalyptic London after an aggression-inducing virus led to the breakdown of society. Murphy failed to return for 2007's 28 Weeks Later, which followed a different cast of characters. But he is executive producer of the 28 Years Later and has a mystery role in the film (that zombie in the 28 Years Later trailer isnt Cillian Murphy, though).Play28 Years Later is set almost three decades after the virus was first unleashed upon the world. Heres the official synopsis:Academy Award-winning director Danny Boyle and Academy Award-nominated writer Alex Garland reunite for 28 Years Later, a terrifying new story set in the world created by 28 Days Later. Its been almost three decades since the rage virus escaped a biological weapons laboratory, and now, still in a ruthlessly enforced quarantine, some have found ways to exist amidst the infected. One such group of survivors lives on a small island connected to the mainland by a single, heavily-defended causeway. When one of the group leaves the island on a mission into the dark heart of the mainland, he discovers secrets, wonders, and horrors that have mutated not only the infected but other survivors as well.In October, Ralph Fiennes discussed his role in 28 Years Later, confirming he plays a good doctor (as opposed to an evil doctor, which definitely exists in the 28 Days universe). He also mentioned the revival was a trilogy and that two films had been shot."Britain is 28 years into this terrible plague of infected people who are violent, rabid humans with a few pockets of uninfected communities," Fiennes said. "And it centers on a young boy who wants to find a doctor to help his dying mother. He leads his mother through this beautiful northern English terrain. But of course, around them hiding in forests and hills and woods are the infected. But he finds a doctor who is a man we might think is going to be weird and odd, but actually is a force for good.After a long period of unavailability online, 28 Days Later was recently made available to rent or buy digitally.Wesley is the UK News Editor for IGN. Find him on Twitter at @wyp100. You can reach Wesley at wesley_yinpoole@ign.com or confidentially at wyp100@proton.me.

HBO has renewed Dune: Prophecy for a second season.The announcement comes ahead of the Season 1 finale, set to debut this Sunday, December 22 at 9pm ET/PT. HBO said that across Max territories, the first episode has seen approximately 15 million viewers.Dune: Prophecy is set 10,000 years before the ascension of Paul Atreides and the events of Denis Villeneuves movies. Inspired by the novel Sisterhood of Dune, Dune: Prophecy follows two Harkonnen sisters who establish the Bene Gesserit, the sisterhood pulling the strings of Frank Herberts expansive science fiction universe.Sarah Aubrey, Head of Max Original Programming, said: Dune: Prophecy has captivated audiences around the globe thanks to the visionary leadership of showrunner and executive producer Alison Schapker, who will continue to guide this grand tale of truth and power. We are incredibly grateful to our partners at Legendary and to our extraordinary cast and crew for their service to the Imperium. Were excited to collaborate with this team again to see what they have in store.Jason Clodfelter, Legendarys President of Television, added: This new season will allow us to continue building out the groundbreaking, epic Dune franchise that has captivated audiences worldwide across its installments. We look forward to continuing our incredible partnership with HBO and are thrilled for Alison Schapker, her team, and the cast and crew who have worked so passionately to bring this world-class source material from Brian Herbert and Kevin J. Anderson to life."Check out IGNs spoiler-free review of Dune: Prophecy's first four episodes for more.Photograph courtesy of HBO.Wesley is the UK News Editor for IGN. Find him on Twitter at @wyp100. You can reach Wesley at wesley_yinpoole@ign.com or confidentially at wyp100@proton.me.

AI video editing based on text prompts is coming to Instagram next year, with the companys head showing off the ability to do things like change your clothing and video background, as well as some funkier stuff like adding animated characters.The company had earlier shown a demo video (below) of completely AI-generated video, stating that it was just a research project, but it seems the first practical application of the tech is arriving sooner than expected Earlier this year, Meta previewed an AI video editing and creation tool known as Movie Gen. This was a brief showreel of video generated by AI using nothing more than text prompts by the user.The company said then that Movie Gen was just research, giving the impression that it wouldnt be available to users for some considerable time. You can the short demo video here:But while creating AI video out of nothing but text may be further down the line, Instagram lead Adam Mosseri showed off a text-based AI video editing feature he says the company is hoping to launch next year.Im super excited about Movie Gen, our early AI research model that will let you change nearly any aspect of your videos with a simple text prompt. Hoping to bring this to Instagram next year. Let me know what you think.The demo shows his cardigan changing into a leather jacket, adding a gold chain around his neck, and his background changing to snowy fields, the Eiffel Tower, and New Yorks Times Square. It also shows the addition of a 3D cartoon cow, and Mosseri himself turning into an Animoji-style character.Theres nothing more specific on timings than a hope for some time in 2025. Image: MetaAdd 9to5Mac to your Google News feed. FTC: We use income earning auto affiliate links. More.Youre reading 9to5Mac experts who break news about Apple and its surrounding ecosystem, day after day. Be sure to check out our homepage for all the latest news, and follow 9to5Mac on Twitter, Facebook, and LinkedIn to stay in the loop. Dont know where to start? Check out our exclusive stories, reviews, how-tos, and subscribe to our YouTube channel

Dec 20, 2024Ravie LakshmananCyber Espionage / MalwareThe Lazarus Group, an infamous threat actor linked to the Democratic People's Republic of Korea (DPRK), has been observed leveraging a "complex infection chain" targeting at least two employees belonging to an unnamed nuclear-related organization within the span of one month in January 2024.The attacks, which culminated in the deployment of a new modular backdoor referred to as CookiePlus, are part of a long-running cyber espionage campaign known as Operation Dream Job, which is also tracked as NukeSped by cybersecurity company Kaspersky. It's known to be active since at least 2020, when it was exposed by ClearSky.These activities often involve targeting developers and employees in various companies, including defense, aerospace, cryptocurrency, and other global sectors, with lucrative job opportunities that ultimately lead to the deployment of malware on their machines."Lazarus is interested in carrying out supply chain attacks as part of the DeathNote campaign, but this is mostly limited to two methods: the first is by sending a malicious document or trojanized PDF viewer that displays the tailored job descriptions to the target," the Russian firm said in an exhaustive analysis."The second is by distributing trojanized remote access tools such as VNC or PuTTY to convince the targets to connect to a specific server for a skills assessment."The latest set of attacks documented by Kaspersky involve the second method, with the adversary making use of a completely revamped infection chain delivering a trojanized VNC utility under the pretext of conducting a skills assessment for IT positions at prominent aerospace and defense companies. It's worth noting that Lazarus Group's use of rogue versions of VNC apps to target nuclear engineers was previously highlighted by the company in October 2023 in its APT trends report for Q3 2023."Lazarus delivered the first archive file to at least two people within the same organization (we'll call them Host A and Host B)," researchers Vasily Berdnikov and Sojun Ryu said. "After a month, they attempted more intensive attacks against the first target."The VNC apps, a trojanized version of TightVNC called "AmazonVNC.exe," are believed to have been distributed in the form of both ISO images and ZIP files. In other cases, a legitimate version of UltraVNC was used to sideload a malicious DLL packed within the ZIP archive.The DLL ("vnclang.dll") serves as a loader for a backdoor dubbed MISTPEN, which was uncovered by Google-owned Mandiant in September 2024. It's tracking the activity cluster under the moniker UNC2970. MISTPEN, for its part, has been found to deliver two additional payloads codenamed RollMid and a new variant of LPEClient.Kaspersky said it also observed the CookieTime malware being deployed on Host A, although the exact method that was used to facilitate it remains unknown. First discovered by the company in September and November 2020, CookieTime is so named for its use of encoded cookie values in HTTP requests to fetch instructions from a command-and-control (C2) server.Further investigation of the attack chain has revealed that the threat actor moved laterally from Host A to another machine (Host C), where CookieTime was again used to drop various payloads between February and June 2024, such as follows -LPEClient, a malware that comes fitted with capabilities to profile compromised hostsServiceChanger, a malware that stops a targeted legitimate service so as to sideload a rogue DLL embedded within it using the executable via DLL side-loadingCharamel Loader, a loader malware that decrypts and loads internal resources like CookieTime, CookiePlus, and ForestTigerCookiePlus, a new plugin-based malicious program that's loaded by both ServiceChanger and Charamel Loader"The difference between each CookiePlus loaded by Charamel Loader and by ServiceChanger is the way it is executed. The former runs as a DLL alone and includes the C2 information in its resources section," the researchers pointed out."The latter fetches what is stored in a separate external file like msado.inc, meaning that CookiePlus has the capability to get a C2 list from both an internal resource and an external file. Otherwise, the behavior is the same."CookiePlus gets its name from the fact that it was disguised as an open-source Notepad++ plugin called ComparePlus when it was detected in the wild for the first time. In the attacks targeting the nuclear-related entity, it has been found to be based on another project named DirectX-Wrappers.The malware serves as a downloader to retrieve a Base64-encoded, RSA-encrypted payload from the C2 server, which is then decoded and deciphered to execute three different shellcodes or a DLL. The shellcodes are equipped with features to collect system information and make the main CookiePlus module sleep for a certain number of minutes.It's suspected that CookiePlus is a successor to MISTPEN owing to behavioral overlaps between the two malware families, including the aspect that both have disguised themselves as Notepad++ plugins."Throughout its history, the Lazarus group has used only a small number of modular malware frameworks such as Mata and Gopuram Loader," Kaspersky said. "The fact that they do introduce new modular malware, such as CookiePlus, suggests that the group is constantly working to improve their arsenal and infection chains to evade detection by security products."The findings come as blockchain intelligence firm Chainalysis revealed that threat actors affiliated with North Korea have stolen $1.34 billion across 47 cryptocurrency hacks in 2024, up from $660.50 million in 2023. This included the May 2024 breach of Japanese cryptocurrency exchange, DMM Bitcoin, which suffered a loss of $305 million at the time."Unfortunately, it appears that the DPRK's crypto attacks are becoming more frequent," the company said. "Notably, attacks between $50 and $100 million, and those above $100 million occurred far more frequently in 2024 than they did in 2023, suggesting that the DPRK is getting better and faster at massive exploits."Found this article interesting? Follow us on Twitter and LinkedIn to read more exclusive content we post.SHARE

Delivering and supporting the right infrastructure for AI, a major challenge in 2024, will remain a top challenge in years to come.

US doctors write billions of prescriptions each year. During 2024, though, one type of drug stood outwonder drugs known as GLP-1 agonists. As of September, one of every 20 prescriptions written for adults was for one of these drugs, according to the health data company Truveta. The drugs, which include Wegovy, Mounjaro, and Victoza, are used to treat diabetes, since they help generate insulin. But their popularity exploded after scientists determined the drugs tell your brain youre not hungry. Without those hunger cues, people find they can lose 10% of their body weight, or even more. During 2024, the drugs popularity hit an all-time high, according to Tricia Rodriguez, a principal applied scientist at Truveta, which studies medical records of 120 million Americans, or about a third of the population. Among adults, 5.4% of all prescriptions in September 2024 were for GLP-1s, Rodriguez says. That is up from 3.5% a year earlier, in 2023, and 1% at the start of 2021. According to Truvetas data, people who get prescriptions for these drugs are younger, whiter, and more likely to be female. In fact, women are twice as likely as men to get a prescription. Yet not everyone whos prescribed the drugs ends up taking them. In fact, Rodriguez says, half the new prescriptions for obesity are going unfilled. Thats very unusual, she says, and could be due to shortages or sticker shock over the cost of the treatment. Many insurers don t cover weight-loss drugs, and the out-of-pocket price can be $1,300 a month, according to USA Today. For most medications, prescribing rates and dispensing rates are pretty much identical, says Rodriguez. But for GLP-1s, we see this gap, which is really unique. It's suggestive that people are really interested in getting these medications, but for whatever reason, they are not always able to. It also means the number of people taking these drugs could go highermaybe much higherif insurers would pay. I don't think that we are at the saturation point, or necessarily nearing the saturation point, says Rodriguez, noting that around 70% of Americans are overweight or obese. Use of the drugs may also grow dramatically if new applications are found. Companies are already exploring whether they can treat addiction, or even Alzheimers. Many of the clues about those potential uses are coming directly out of peoples medical records. Because so many people are on the drugs, it means researchers like Rodriguez have a gold mine to sift through for signs of how use of the drugs is affecting other health problems. Because we have so many patients that are on these medications, you're certainly likely to have a good number that also have all of these other conditions, she says. One of the things we're excited about is: How can real-world data help accelerate how quickly we can understand those? Here are some of the new uses of GLP-1 drugs that are being explored, based on hints from real-world patient records. Alzheimers disease This year, researchers poking through records of a million people found that taking semaglutide (sold as Wegovy and Ozempic) was associated with a 40% to 70% lower chance of an Alzheimers diagnosis. It's still a guess why the drugs might be helping (or whether they really do), but large international studies are underway to follow up on the lead. Doctors are recruiting people with early Alzheimers in more than 30 countries who will take either a placebo or semaglutide for two years. Then well see how much their dementia has progressed. Addiction The anecdotes are everywhere: A person on a weight-loss drug finds hunger isnt the only craving that seems to stop. Those are the types of clues Eli Lillys CEO, David Ricks, says his company will pursue next year, testing whether its GLP-1 drug, tirzepatide (called Mounjaro for diabetes treatment, and Zepbound for weight loss), could help with addiction to alcohol, nicotine, and other things we dont think about [as being] connected to weight. In comments he made in December, Ricks said the drugs might be anti-hedonicsmeaning they counteract our hedonistic pursuit of pleasure, be it from food, alcohol, or drugs. A study this year mining digital health records found that opioid addicts taking the drugs were about half as likely to have had an overdose. Sleep apnea This idea goes back a ways, including to a 2015 case study of a 260-pound man with diabetes and sleep apnea. When he went on the drug liraglutide, doctors noticed that his sleeping improved. In sleep apnea, a person gasps for air at nightits annoying and, with time, causes health problems. This year, Eli Lilly published a study in the New England Journal of Medicine on its drugtirzepatide , finding that it caused a 50% decrease in breathing interruption in overweight patients with sleep apnea. Longevity This year, the U.S. Food and Drug Administration approved Wegovy as a cardiovascular medicine, after researchers showed the drugs could reduce heart attack and stroke in overweight people. But that wasnt all. The study, involving 17,000 people, found that the drug reduced the overall chance someone would die for any reason (known as all-cause mortality) by 19%. That now has aging researchers paying attention. This year they named Wegovy, and drugs like it, among their the top four candidates for a general life-extension drug.

Networks programmed directly into computer chip hardware can identify images faster, and use much less energy, than the traditional neural networks that underpin most modern AI systems. Thats according to work presented at a leading machine learning conference in Vancouver last week. Neural networks, from GPT-4 to Stable Diffusion, are built by wiring together perceptrons, which are highly simplified simulations of the neurons in our brains. In very large numbers, perceptrons are powerful, but they also consume enormous volumes of energyso much that Microsoft has penned a deal that will reopen Three Mile Island to power its AI advancements. Part of the trouble is that perceptrons are just software abstractionsrunning a perceptron network on a GPU requires translating that network into the language of hardware, which takes time and energy. Building a network directly from hardware components does away with a lot of those costs. One day, they could even be built directly into chips used in smartphones and other devices, dramatically reducing the need to send data to and from servers. Felix Petersen, who did this work as a postdoctoral researcher at Stanford University, has a strategy for making that happen. He designed networks composed of logic gates, which are some of the basic building blocks of computer chips. Made up of a few transistors apiece, logic gates accept two bits1s or 0sas inputs and, according to a rule determined by their specific pattern of transistors, output a single bit. Just like perceptrons, logic gates can be chained up into networks. And running logic-gate networks is cheap, fast, and easy: in his talk at the Neural Information Processing Systems (NeurIPS) conference, Petersen said that they consume less energy than perceptron networks by a factor of hundreds of thousands. Logic-gate networks dont perform nearly as well as traditional neural networks on tasks like image labeling. But the approachs speed and efficiency make it promising, according to Zhiru Zhang, a professor of electrical and computer engineering at Cornell University. If we can close the gap, then this could potentially open up a lot of possibilities on this edge of machine learning, he says. Petersen didnt go looking for ways to build energy-efficient AI networks. He came to logic gates through an interest in differentiable relaxations, or strategies for wrangling certain classes of mathematical problems into a form that calculus can solve. It really started off as a mathematical and methodological curiosity, he says. Backpropagation, the training algorithm that made the deep-learning revolution possible, was an obvious use case for this approach. Because backpropagation runs on calculus, it cant be used directly to train logic-gate networks. Logic gates only work with 0s and 1s, and calculus demands answers about all the fractions in between. Petersen devised a way to relax logic-gate networks enough for backpropagation by creating functions that work like logic gates on 0s and 1s but also give answers for intermediate values. He ran simulated networks with those gates through training and then converted the relaxed logic-gate network back into something that he could implement in computer hardware. One challenge with this approach is that training the relaxed networks is tough. Each node in the network could end up as any one of 16 different logic gates, and the 16 probabilities associated with each of those gates must be kept track of and continually adjusted. That takes a huge amount of time and energyduring his NeurIPS talk, Petersen said that training his networks takes hundreds of times longer than training conventional neural networks on GPUs. At universities, which cant afford to amass hundreds of thousands of GPUs, that amount of GPU time can be tough to swingPetersen developed these networks, in collaboration with his colleagues, at Stanford University and the University of Konstanz. It definitely makes the research tremendously hard, he says. Once the network has been trained, though, things get way, way cheaper. Petersen compared his logic-gate networks with a cohort of other ultra-efficient networks, such as binary neural networks, which use simplified perceptrons that can process only binary values. The logic-gate networks did just as well as these other efficient methods at classifying images in the CIFAR-10 data set, which includes 10 different categories of low-resolution pictures, from frog to truck. It achieved this with fewer than a tenth of the logic gates required by those other methods, and in less than a thousandth of the time. Petersen tested his networks using programmable computer chips called FPGAs, which can be used to emulate many different potential patterns of logic gates; implementing the networks in non-programmable ASIC chips would reduce costs even further, because programmable chips need to use more components in order to achieve their flexibility. Farinaz Koushanfar, a professor of electrical and computer engineering at the University of California, San Diego, says she isnt convinced that logic-gate networks will be able to perform when faced with more realistic problems. Its a cute idea, but Im not sure how well it scales, she says. She notes that the logic-gate networks can only be trained approximately, via the relaxation strategy, and approximations can fail. That hasnt caused issues yet, but Koushanfar says that it could prove more problematic as the networks grow. Nevertheless, Petersen is ambitious. He plans to continue pushing the abilities of his logic-gate networks, and he hopes, eventually, to create what he calls a hardware foundation model. A powerful, general-purpose logic-gate network for vision could be mass-produced directly on computer chips, and those chips could be integrated into devices like personal phones and computers. That could reap enormous energy benefits, Petersen says. If those networks could effectively reconstruct photos and videos from low-resolution information, for example, then far less data would need to be sent between servers and personal devices. Petersen acknowledges that logic-gate networks will never compete with traditional neural networks on performance, but that isnt his goal. Making something that works, and that is as efficient as possible, should be enough. It wont be the best model, he says. But it should be the cheapest.