Microsofts third consecutive annual Xbox Developer_Direct presentation is around the corner, and unsurprisingly, its generating plenty of buzz about what audiences will be shown. Microsoft has confirmed that it will be showcasingSouth of Midnight, DOOM: The Dark Ages,andClair Obscur: Expedition 33,and that that trio will be joined by another studios brand new game in a surprise location.Now, a report published by Windows Central may have potentially shed some light on that mystery fourth title. Allegedly, Microsoft will also be unveiling a new entry in a legendary Japanese franchise at the Developer Direct presentation. Precise details havent been shared, but apparently, the IP in question has decades of history, and the new announcement will make a lot of fans happy. Take from that what you will.Persona? Resident Evil?Perhaps something dormant likeNinja Gaiden? One can hope.The Xbox Developer Direct showcase is set for January 23 at 10 AM Pacific / 1 PM Eastern, so well know more soon enough. Stay tuned for all the updates.

Two newly discovered, spiky creatures from the Silurian periodnicknamed Punk and Emoare helping scientists reveal the early history of mollusks. Mark Sutton, Imperial College London. Altered into GIFMove aside Fall Out Boy and My Chemical Romancetwo new rockstars have just entered the punk and emo scene.Meet Punk ferox and Emo vorticaudum, two newly revealed fossils rocking scientists previously held theories about a group of ancient mollusks. The pair of more than 430-million-year-old creatures, described in the journal Nature last week, suggests these long-extinct species are not as primitive as paleontologists once thought.The names Punk and Emo were actually our initial pet names for these ancient mollusks, inspired by some of their unique features and individuality, Mark Sutton, a paleontologist at Imperial College London who led the study, explains in a statement. Punk in particular, with its spiky appearance, clearly resembles a rebellious punk rockerand we thought Emo complemented it well.Mollusks are the second largest animal phylum in the world, behind only Arthropods (a group that includes insects, spiders and crustaceans). Scientists suggest the Earth might host up to 200,000 species of mollusks. But this vast array boils down to two main groups: the conchiferans, which includes animals like snails, clams and octopuses, and the aculiferans, a weird and wacky group of spiny things including obscure, weird molluscan worms, as Sutton tells the New York Times Kate Golembiewski.Punk and Emo fall into the latter classand they certainly are weird and spiny. Both Punk and Emo are described as worm-like mollusks with long spines and smooth undersides that suggest they inhabited the seafloor. Punk had gills and a broad foot with ridges on the bottom, making it stand out from living species. Emo, meanwhile, had two small shells and a compressed body. Specifically, Sutton found Punks spikes to closely resemble classic punk hairstyles, and he thought Emos more downward-pointing spines call to mind emo bangs, per the New York Times. A labeled 3D model of Punk (top) and Emo (bottom) Mark Sutton, Imperial College LondonWhile scientists are still unsure how Punk would have moved, they discovered Emo in a folded posture that hints to an inchworm-like locomotion, perhaps aided by its spines. However, the spikes are probably mostly protective, Sutton tells New Scientists Chris Simms, though he adds that they could have formed to rid the animals body of accumulated calciumand both explanations could be true. The mollusks were small, at about 0.79 inches long.Aculiferans like Punk and Emo were long considered to be much more limited in form than the more familiar mollusks, as Derek Briggs, a paleontologist at Yale University and a co-author of the study, says in a statement. Now, these stunning fossils show their group was remarkably diverse in the distant past, compared to the present day, he adds.Briggs and his colleagues discovered Punk and Emo preserved in 3D within stones from a site known as the Herefordshire Lagersttte in the United Kingdom, which dates back to the middle of the Silurian period (about 420 million to 443 million years ago). The fossils were so fragile that the scientists couldnt simply split the stones in half, and their density was so similar to the surrounding rock that even CT scans wouldnt have offered a clear picture of the animals. Instead, they sliced the fossils up by hand and took a picture of each tiny section.We grind away a slice at a time, take a photo, repeat at 20-micron intervals or so, explains Sutton to the New York Times. Though this means the fossils were destroyedturned to dust by the end of the processthe scientists were able to combine the images into detailed 3D representations of the mollusks.I really like the names, and they definitely fit the bill for these spiky mollusks, Luke Parry, a paleontologist at the University of Oxford in England who was not involved in the study, tells New Scientist. Fossil mollusks that preserve soft tissues like this are incredibly rare, and so seeing what these unusual ancient animals looked like in 3D is just spectacular. This site in Herefordshire appears to be a treasure trove, like a wormy mollusk Pompeii.Get the latest stories in your inbox every weekday.Filed Under: Animals, Cool Finds, Evolution, Fossils, Mollusks, Worms, Sponges, Starfish, Music, New Research, Paleontology

Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn MoreThe era of reasoning AI is well underway.After OpenAI once again kickstarted an AI revolution with its o1 reasoning model introduced back in September 2024 which takes longer to answer questions but with the payoff of higher performance, especially on complex, multi-step problems in math and science the commercial AI field has been flooded with copycats and competitors.Theres DeepSeeks R1, Google Gemini 2 Flash Thinking, and just today, LlamaV-o1, all of which seek to offer similar built-in reasoning to OpenAIs new o1 and upcoming o3 model families. These models engage in chain-of-thought (CoT) prompting or self-prompting forcing them to reflect on their analysis midstream, double back, check over their own work and ultimately arrive at a better answer than just shooting it out of their embeddings as fast as possible, as other large language models (LLMs) do.Yet the high cost of o1 and o1-mini ($15.00/1M input tokens vs. $1.25/1M input tokens for GPT-4o on OpenAIs API) has caused some to balk at the supposed performance gains. Is it really worth paying 12X as much as the typical, state-of-the-art LLM?As it turns out, there are a growing number of converts but the key to unlocking reasoning models true value may lie in the user prompting them differently.Shawn Wang (founder of AI news service Smol) featured on his Substack over the weekend a guest post from Ben Hylak, the former Apple Inc., interface designer for visionOS (which powers the Vision Pro spatial computing headset). The post has gone viral as it convincingly explains how Hylak prompts OpenAIs o1 model to receive incredibly valuable outputs (for him).In short, instead of the human user writing prompts for the o1 model, they should think about writing briefs, or more detailed explanations that include lots of context up-front about what the user wants the model to output, who the user is and what format in which they want the model to output information for them.As Hylak writes on Substack:With most models, weve been trained to tell the modelhowwe want it to answer us. e.g. You are an expert software engineer. Think slowly and carefullyThis is the opposite of how Ive found success with o1. I dont instruct it on thehow only thewhat. Then let o1 take over and plan and resolve its own steps. This is what the autonomous reasoning is for, and can actually be much faster than if you were to manually review and chat as the human in the loop.Hylak also includes a great annotated screenshot of an example prompt for o1 that produced a useful results for a list of hikes:This blog post was so helpful, OpenAIs own president and co-founder Greg Brockman re-shared it on his X account with the message: o1 is a different kind of model. Great performance requires using it in a new way relative to standard chat models.I tried it myself on my recurring quest to learn to speak fluent Spanish and here was the result, for those curious. Perhaps not as impressive as Hylaks well-constructed prompt and response, but definitely showing strong potential.Separately, even when it comes to non-reasoning LLMs such as Claude 3.5 Sonnet, there may be room for regular users to improve their prompting to get better, less constrained results.As Louis Arge, former Teton.ai engineer and current creator of neuromodulation device openFUS, wrote on X, one trick ive discovered is that LLMs trust their own prompts more than my prompts, and provided an example of how he convinced Claude to be less of a coward by first trigger[ing] a fight with him over its outputs.All of which goes to show that prompt engineering remains a valuable skill as the AI era wears on.Daily insights on business use cases with VB DailyIf you want to impress your boss, VB Daily has you covered. We give you the inside scoop on what companies are doing with generative AI, from regulatory shifts to practical deployments, so you can share insights for maximum ROI.Read our Privacy PolicyThanks for subscribing. Check out more VB newsletters here.An error occured.

TikTokers are coping with the apps potential ban with an unusual trend: by bidding farewell to their personal Chinese spy. The trend, which pokes fun at security concerns surrounding the app, has users thanking their spy for surveilling them and filling their For You page with entertaining content, while others proclaim that theyd rather share their data directly with the Chinese government than switch to Instagram Reels or YouTube Shorts.One post, which garnered more than 1.5 million likes, depicts an emotional scene from Squid Game with the caption, Me saying goodbye to my Chinese spy on the 19th (He perfected my algorithm). Other TikTokers are speaking and singing in Chinese, while some pretend to be the spies powering individual algorithms.It is a great honor to spy on you for the last few years, TikToker yanxiao1003 says in a video. I wish you all have a great life in the future... Laura from California, you shouldnt drink that much Coca-Cola, its bad for your health.RelatedFor years, TikTok has faced scrutiny over the ties of its parent company, ByteDance, to China, with lawmakers calling it a national security threat and accusing the app of sending US user data to China. But now the platforms ban is imminent, with the Supreme Court set to hand down a decision on whether to uphold the law that would either ban TikTok or force ByteDance to divest it by January 19th.

The rapid advancements in artificial intelligence have opened new possibilities, but the associated costs often limit who can benefit from these technologies. Large-scale models like GPT-4 and OpenAIs o1 have demonstrated impressive reasoning and language capabilities, but their development and training remain financially and computationally burdensome. This creates barriers for smaller organizations, academic institutions, and independent researchers. Moreover, the closed-source nature of many advanced models restricts broader access, limiting opportunities for collaborative innovation. This raises a critical question: How can cutting-edge AI technologies become accessible to a wider audience without compromising quality?In response to these challenges, researchers at UC Berkeley have introduced Sky-T1-32B, a reasoning-focused language model that is both open-source and cost-efficient. Sky-T1s standout feature is its affordabilitythe model can be trained for less than $450. With 32 billion parameters, the model is carefully designed to balance computational efficiency with robust performance. The development process emphasizes practical and efficient methodologies, including optimized data scaling and innovative training pipelines, enabling it to compete with larger, more resource-intensive models.Sky-T1s open-source nature fosters inclusivity in AI research and development. By making the models architecture and training process freely available, the UC Berkeley team aims to empower researchers and developers worldwide to customize and apply Sky-T1 to diverse use cases. This initiative addresses long-standing limitations posed by proprietary systems and paves the way for collaborative advancements in AI.Technical Insights and Key BenefitsSky-T1 achieves its cost efficiency through a series of carefully implemented technical strategies. The models training process relies on optimized data scaling and parameter-efficient techniques, ensuring effective resource utilization. Methods like sparse computation and low-rank adaptation (LoRA) reduce the models memory and compute requirements without compromising performance. Furthermore, its architecture incorporates reasoning-centric pretraining, enhancing its ability to handle logical inference and complex problem-solving tasks.The key benefits of Sky-T1 include:Affordability: Training costs under $450 make Sky-T1 accessible to a broader range of users, including smaller institutions and individual developers.Open Access: The open-source design encourages collaboration and customization, breaking down barriers to innovation.Reasoning Optimization: Unlike general-purpose LLMs, Sky-T1 is fine-tuned for reasoning tasks, making it highly effective in education, research, and automated decision-making.Sustainability: The models reduced computational requirements align with environmental sustainability goals by minimizing energy consumption.Performance Evaluation and InsightsSky-T1 has been tested against established benchmarks such as Math500, AIME, and Livebench, which evaluate reasoning and problem-solving capabilities. On medium and hard tasks within these benchmarks, Sky-T1 outperforms OpenAIs o1, a notable competitor in reasoning-focused AI. For instance, on Math500a benchmark for mathematical reasoningSky-T1 demonstrates superior accuracy while requiring fewer computational resources.The models adaptability is another significant achievement. Despite its relatively modest size, Sky-T1 generalizes well across a variety of reasoning tasks. This versatility is attributed to its high-quality pretraining data and a deliberate focus on reasoning-centric objectives. Additionally, the training process, which requires just 19 hours, highlights the feasibility of developing high-performance models quickly and cost-effectively.Conclusion: A Path Toward Inclusive AIUC Berkeleys Sky-T1 model represents a meaningful step toward making advanced AI technologies more accessible and equitable. By significantly reducing the cost of training and offering an open-source framework, Sky-T1 has the potential to transform how AI is developed and deployed. Its performance on reasoning benchmarks demonstrates that affordability does not necessitate a trade-off in quality. As Sky-T1 gains traction among researchers and developers, it may inspire a wave of innovation that extends AIs benefits to underserved sectors and communities. In this sense, Sky-T1 is more than a technological achievement; its a blueprint for a more inclusive AI future.Check out the Model on Hugging Face, Details, 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 our65k+ ML SubReddit. Asif RazzaqAsif Razzaq is the CEO of Marktechpost Media Inc.. As a visionary entrepreneur and engineer, Asif is committed to harnessing the potential of Artificial Intelligence for social good. His most recent endeavor is the launch of an Artificial Intelligence Media Platform, Marktechpost, which stands out for its in-depth coverage of machine learning and deep learning news that is both technically sound and easily understandable by a wide audience. The platform boasts of over 2 million monthly views, illustrating its popularity among audiences. Meet 'Height':The only autonomous project management tool (Sponsored)

In 2023, Winnie-the-Pooh: Blood and Honey, with a budget of just $50,000, grossed over $5 million at the box office. Now, it seems just about any childhood story that enters the public domain is destined for at least one gory adaptation in the Twisted Child Universe. The newest addition to the TCU, Peter Pans Neverland Nightmare, introduces a bloodthirsty spin on Tinker Bell and the boy who never grows up. In his review for IGN, critic Matt Donato describes how director Scott Jeffrey strives to accentuate his main characters arrested adolescence by turning the boy who never grew up into a serial killer and kidnapper keeping him more grounded than the typical pirates-and-fairies image of Peter Pan. While this is the first new addition to the Twisted Universe since the Blood and Honey movies, more twisted childhood tales are on the way, eventually leading to a multiverse-style crossover project.If youre hoping to catch Peter Pans Neverland Nightmare in theaters or want to stream other movies in the Twisted Child Universe, check out the details below. PlayHow to Watch Peterpans Neverland Nightmare - Showtimes and Streaming Release Date Peter Pans Neverland Nightmare just released in theaters, with limited showtimes on January 13, 14, and 15. You can check out showtimes near you at the main theater links below: Neverland Nightmare Streaming Release DateNot a whole lot has been confirmed about the streaming release for Neverland Nightmare, but we can make some predictions. Like Neverland Nightmare, Winnie-the-Pooh: Blood and Honey 2 had a limited three-day theatrical release. It arrived on digital marketplaces soon after it left theaters in March, then arrived on Peacock in October. Based on that, Neverland Nightmare should be available to rent on demand by the end of January and could arrive on Peacock by July. Well be sure to keep this article updated with confirmation on the movies streaming destination and release date. What Is Peter Pans Neverland Nightmare About? The latest "Twisted Childhood" movie introduces a bloodthirsty interpretation of the Peter Pan story originally created by J. M. Barrie. Heres the movies official synopsis: Wendy Darling encounters a twisted Tinkerbell as she tries to rescue her brother, Michael, from the evil clutches of Peter Pan.Where to Stream the Twisted Child Universe Peacock PremiumSee itWinnie-the-Pooh: Blood and Honey Stream: Peacock Rent/Buy: Prime VideoIGN's ReviewWinnie-the-Pooh: Blood and Honey 2 Stream: Peacock Rent/Buy: Prime VideoIGN's ReviewPeter Pan's Neverland Nightmare Cast Peter Pans Neverland Nightmare was written and directed by Scott Jeffrey, who portrayed Christopher Robin in Blood and Honey 2. Neverland Nightmare stars the following the cast: Martin Portlock as Peter PanMegan Placito as Wendy DarlingKit Green as Tinker BellPeter DeSouza-Feighoney as Michael DarlingCharity Kase as Captain James HookTeresa Banham as Mary DarlingNicholas Woodeson as StevenKierston Wareing as RoxyOlumide Olorunfemi as Tiger LilyCampbell Wallace as John Darling

Apple has a huge lineup of products coming in 2025, including three new Home products, six devices getting rare updates, five new iPhones, and more. Based on the reporting of Mark Gurman and Ming-Chi Kuo especially, heres a full lineup of whats coming.Pretty soon, well start to get the first Apple product launches of the new year. Here are the first products expected to arrive.M4 MacBook Air: Better battery, camera, and external display support.iPhone SE 4: OLED edge-to-edge display, AI features, and more.iPad 11: Upgraded chip to support Apple Intelligence.New iPad Air: Small revision to the M3 or M4 chip.HomePad: Brand new smart home display.Summer 2025: Light on hardware, heavy on softwareApples summer is usually a quieter time for hardware launches, as WWDC kicks off a major focus on software instead. Nonetheless, were expecting the following:AirTag 2: Better range and privacy features.M4 Mac Studio: First M4 Ultra chip.M4 Mac Pro: Largely unknown, likely unveiled at WWDC but launching later.Fall 2025: Apples biggest launch seasonThis is the biggest time of the year for Apple, with major launches usually spread across September and October.iPhone 17: 24MP front camera, higher refresh rate display.iPhone 17 Air: Ultra-thin new design, form over function.iPhone 17 Pro: Better chip, more RAM, upgraded cameras.iPhone 17 Pro Max: All of the above, with a smaller Dynamic Island.Apple Watch Ultra 3: Satellite connectivity, high blood pressure detection, 5G.Apple Watch Series 11: New health sensors likely.Apple Watch SE 3: Fresh look, more modern internals, possibly plastic.AirPods Pro 3: H3 chip, new design, better ANC.New Apple TV 4K: Possibly AI support.HomePod mini 2: In-house Wi-Fi chip.M5 iPad Pro: Spec bump update.M5 MacBook Pro: Spec bump update.Potential surprises and Apples 2025 wrap-upThere are always ways Apple could surprise us with new products not listed above. For example, it would be a great year to launch a new Studio Display alongside the Mac Studio and Mac Pro. Additionally, an M5 Vision Pro is expected to go into production in the second half of the year, which could mean a 2025 launch or perhaps early 2026.Possible surprises aside, it looks like an especially busy year ahead for Apple hardware releases.Which new Apple products are you most looking forward to in 2025? Let us know in the comments.Best iPhone accessoriesAdd 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

AI companies continue to reach for tortuous narratives to justify stuffing AI tech into virtually every aspect of our daily lives.Take Mikey Shulman, the CEO of AI music generator company Suno AI, who had some choice words about how he sees the creative process when it comes to making music during an appearance on the "20VC" podcast flagged by 404 Media."Its not really enjoyable to make music now," he said. "It takes a lot of time, it takes a lot of practice, you need to get really good at an instrument or really good at a piece of production software.""I think the majority of people dont enjoy the majority of the time they spend making music," he added.Shulman's comments are astonishingly self-serving, as that's exactly the kind of audience Suno AI is hoping to attract. And in reality, of course, lots of people amateurs and professionals alike love playing music.If it were up to Shulman, it seems, music would devolve into an amalgamation of often copyright-infringing AI slop, allowing "artists" to create tracks with a click of a button.In short, arguing that the majority of people no longer enjoy making music is symptomatic of an industry hellbent on eliminating as much of the human element from creative endeavors as possible a dystopian vision,critics charge, of art as a whole.According to Shulman, Suno AI is aiming to make making music "ten percent faster" and "ten percent easier.""If you want to impact the way a billion people experience music you have to build something for a billion people," he told the 20VC podcast.But whether the resulting, AI-generated music will give listeners the same kind of satisfaction as creating music themselves remains debatable.Sure, creating your own bespoke track by typing some words into an AI music generator is a fun trick, but relegating almost the entire creative process to an algorithm leaves plenty to be desired.Suno AI has also been caught up in a number of copyright lawsuits, with the Recording Industry Association of America accusing the startup of training its AI on copyrighted recordings without permission.In other words, Shulman doesn't just want people to give up agency in the creative process, but his company may be ripping off the work of human artists as well.Shulman's comments also highlight a nagging debate surrounding the use of these kinds of tools: can we even claim ownership over the resulting music? Is taking shortcuts with the help of AI really what human music artists should strive for? Could AI-generated music even be called "art?"Users on social media were almost unanimously appalled by Shulman's remarks on the podcast, calling his stance that almost nobody enjoys making music an "absurd mindset.""It's very revealing," one X user tweeted in response. "The solutionist language just doesn't make sense for art, or for other things where the work is the enjoyment, but the tech companies seem stuck in their story.""What a dystopian future," one YouTube commenter wrote. "And lack of self-criticality is especially disconcerting."Share This Article

Jan 14, 2025Ravie LakshmananVulnerability / CybersecurityThe U.S. Cybersecurity and Infrastructure Security Agency (CISA) on Monday added a second security flaw impacting BeyondTrust Privileged Remote Access (PRA) and Remote Support (RS) products to the Known Exploited Vulnerabilities (KEV) catalog, citing evidence of active exploitation in the wild.The vulnerability in question is CVE-2024-12686 (CVSS score: 6.6), a medium-severity bug that could allow an attacker with existing administrative privileges to inject commands and run as a site user."BeyondTrust Privileged Remote Access (PRA) and Remote Support (RS) contain an OS command injection vulnerability that can be exploited by an attacker with existing administrative privileges to upload a malicious file," CISA said."Successful exploitation of this vulnerability can allow a remote attacker to execute underlying operating system commands within the context of the site user."The addition of CVE-2024-12686 to the KEV catalog comes nearly a month after it added another critical security flaw impacting the same product (CVE-2024-12356, CVSS score: 9.8) that could also lead to the execution of arbitrary commands.BeyondTrust said both vulnerabilities were discovered as part of its investigation into a cyber incident in early December 2024 that involved malicious actors leveraging a compromised Remote Support SaaS API key to breach some of the instances, and reset passwords for local application accounts.Although the API key has since been revoked, the exact manner in which the key was compromised remains unknown as yet. It's suspected that the threat actors exploited the two flaws as zero-days to break into BeyondTrust systems.Earlier this month, the U.S. Treasury Department revealed its network was breached using the compromised API key in what it said was a "major cybersecurity incident." The hack has been pinned on a Chinese state-sponsored group called Silk Typhoon (aka Hafnium).The threat actors are believed to have specifically targeted the Treasury's Office of Foreign Assets Control (OFAC), Office of Financial Research, and the Committee on Foreign Investment in the United States (CFIUS), according to multiple reports from the Washington Post and CNN.Also added to the KEV catalog is a now-patched critical security vulnerability affecting Qlik Sense (CVE-2023-48365, CVSS score: 9.9) that allows an attacker to escalate privileges and execute HTTP requests on the backend server hosting the software.It's worth noting that the security flaw has been actively exploited in the past by the Cactus ransomware group. Federal agencies are required to apply the necessary patches by February 3, 2024, to secure their networks against active threats.Found this article interesting? Follow us on Twitter and LinkedIn to read more exclusive content we post.SHARE

If you've ever wondered whether the wonder product retinol is right for you, you're not alone. I've spent hours scouring skin care Reddit to work out if it's something I should use, as well as which version is right for me. And then when I have used it, whether it's working the way it should.I'm one of the nine out of 10 women who feel frustrated about skin care and can't work out which products are best for me, according to cosmetics giant L'Oreal. There's a "joy" in experimenting with makeup, but experimenting with skin care is "frustrating," says Guive Balooch, the company's global managing director of augmented beauty and open innovation."Part of the problem with that is the fact that we're doing a lot of auditioning without facts and science behind what our skin individually actually needs," Balooch says. Last year, I tried out some of L'Oreal's cutting-edge tech products at Vivatech in Paris, including Kiehl's Derma-Reader, an imaging machine that can examine your skin and make product recommendations. The Derma-Reader was fascinating to experiment with, but the technology looks at only the top surface level of your skin, and also recommends only Kiehl's products as a treatment, which may not be the best for you.At CES 2025 this year, the company unveiled a new device, the Cell BioPrint, a collaboration with Korean beauty tech company NanoEnTek. It can take a reading from under our skin and read the biomarkers to tell us whether we should be responsive to certain ingredients like retinol, for example and then match our skin with appropriate products. Watch this: I Gave L'Oral My Cells to Know What Products Are Good for My Skin (And So Can You!) 02:32 I was excited to try the tech, as I've experimented with retinol a little over the years and haven't been able to work out whether it's for me. Retinol a form of vitamin A is affordable, available over the counter and promises to smooth fine lines and wrinkles, shrink enlarged pores and treat acne. But it also has its downsides: namely that your skin can take a long time to adjust to it, and in the meantime, it can become dry and flaky, or it can cause an acne purge.I've twice abandoned retinol during an ongoing acne purge, often referred to as the retinol uglies. But as L'Oreal's imaging tech has told me, my skin's age is three years older than my actual age and if I can tolerate retinol, the anti-aging effects could be well worth it.What the Cell BioPrint revealed to me Upgrade your inbox Get cnet insider From talking fridges to iPhones, our experts are here to help make the world a little less complicated. My Cell BioPrint experience started with dabbing my skin multiple times with a sticker to pick up skin cells. The sticker went into a cartridge, and it took about 1 minute to strip out everything but the proteins from my skin. The liquid within the cartridge containing the proteins then got squeezed from a dropper onto a test strip and inserted into the tabletop Cell BioPrint machine, which took around 90 seconds to read my biomarkers.The good news: L'Oreal's tech told me that my skin has a high responsiveness to retinol, with a score of 75 out of 100. If I can tolerate it, I might be able to improve my pores, oiliness and barrier function, all of which L'Oreal identified as "needing attention." Along with retinol, it gave me recommendations for a SkinCeuticals salicylic acid serum and a zinc PCA IT cosmetics cream.I really appreciated the scientific analysis of my skin, which made me determined to persevere with retinol the next time I feel like quitting, even though I might need to start at a lower concentration than I've previously used. I also liked that the service, which is designed to provide in-store skin analysis, was able to tap into L'Oreal's vast library of skincare products, rather than solutions just from one brand."Today, we have so many choices, and what we need to understand is, what are the right choices for our biology and our needs?" Balooch said. "A lot of the products take time to work. So this really is seeing the invisible and taking the guessing game out and giving you the most precise products." CES 2025: See the 35 Coolest Tech Products We Can't Shake See all photos