General manager of the Diablo franchise at Blizzard, Rod Fergusson, has spoken about the future of Diablo 4, as well as the franchise as a whole going forward. Speaking to IGN after his talk during DICE Summit 2025, Fergusson revealed that Diablo 4 suffering from something like the catastrophic launch of Diablo 3 with the infamous Error 37 would be disastrous for the game.Fergusson also spoke about the content plan for Diablo 4, and whether or not it has been planned as a game that could potentially go on forever. He spoke about the challenges with doing something like that, citing Bungies attempt with the Destiny franchise.We want it to be around for years, said Fergusson. I dont know if its eternal. I think Destiny tried and did that like, This is a ten-year game, and then they quickly were not. We want people to see the road ahead, because we know that, to play a Diablo game, youre probably putting hundreds of hours in and we want people to know that were respecting their time and that were not just here and gone.When asked about the cadence of expansion releases, Fergusson said that, right now, the studio cant make any promises one way or another. He said that the studios process is to first get to a place internally where it can decide on working on an expansion before formally announcing one.I learned my lesson about calling the shot too early, he said. So I think giving, Hey, you can relax for the next 11 months, is about as far as I want to go in this right nowWere not at a place where we want to put a stake in the ground and call it. Because were still building it and learning from it Our process is generally that we get internally to a place where we have the certainty to make a call, and then you get to a place where you publicly want to make that call. Were not at the public part yet.For context, Diablo 4s expansion, Vessel of Hatred, was released around 18 months after the base game was first released. The expansion brought with it a host of new features to Diablo 4, including an entirely new region and new playable class.The region brought to the game by the expansion, Nahantu, was features a host of throwbacks to seminal Diablo 2, including an explorable version of the city of Kurast. The Spiritborn class was also a big addition to the game; the class is able to wield two of the new weapon types, and uses animal guardians that grant unique benefits.Diablo 4 is currently in the midst of its Season of Witchcraft, which revolves around some of the heads from the Tree of Whispers having gone missing. The season introduces new Witchcraft Powers across three schools: Eldritch, Psyche, and Growth and Decay.For more details about Diablo 4, check out our review of the original. Also check out our review of Vessel of Hatred to learn more about the expansion.

Capcom promotes Monster Hunter producer to chief product officer in charge of all developmentRyozo Tsujimoto assumes new role from April 1Image credit: Capcom News by Vikki Blake Contributor Published on March 3, 2025 Capcom's current executive corporate officer and chief product officer, Yoichi Egawa, will step down on April 1, with Monster Hunter producer Ryozo Tsujimoto assuming the latter role from that same date.President and COO Haruhiro Tsujimoto will take over Egawa's role of overseeing Capcom's Pachinko and Pachislo business divisions.Ryozo Tsujimoto - who currently serves as director and executive corporate officer deputy in charge of development divisions - has been with the Japanese megacorp since joining as a graduate in April 1996.Kenkichi Nomura is similarly stepping down from his position as executive VP, with Yoshinobu Sasahara and Satoshi Arai moving to new positions of executive corporate officer and managing corporate officer, respectively.Both Egawa and Nomura will continue to serve on Capcom's board as directors.Capcom released its financial results for the nine months ending December 31, 2024, last month, reporting net sales and operating income "in-line" with its fiscal plan, despite a 16.3 percent year-on-year drop in net sales to 88.9 billion ($572.7 million) and a 33% fall in net income.

Former NetEase director self-funds new studio to ensure "management protects creators"Founder Kenji Ozawa will assume any project failure rather than pass the blame onto dev teamImage credit: Studio Sasanqua News by Vikki Blake Contributor Published on March 3, 2025 Months after leaving NetEase-owned Ouka Studios - which closed down last year shortly after Visions of Mana was published by Square Enix - co-director Kenji Ozawa has established Studio Sasanqua.Talking to Automaton, Ozawa stressed the new studio will ensure "management [...] protect[s] creators," motivating both developers and development in ways to make staff feel safe and valued.Furthermore, by self-funding his venture, Ozawa reportedly said he was prepared to assume any project failure rather than pass the blame onto the development team itself, as prior experience had shown that some studios placed too much responsibility and blame on the shoulders of individual developers rather than mismanagement.Initially, the team will focus on small-scale projects developed in Unreal Engine, with the goal of transitioning to original IP for PC and consoles.Visions of Mana's other co-director, Ryosuke Yoshida, joined Square Enix in December.

TechTarget and Informa Techs Digital Business Combine.TechTarget and InformaTechTarget and Informa Techs Digital Business Combine.Together, we power an unparalleled network of 220+ online properties covering 10,000+ granular topics, serving an audience of 50+ million professionals with original, objective content from trusted sources. We help you gain critical insights and make more informed decisions across your business priorities.Activision Blizzard is using generative AI to hawk games that don't existActivision Blizzard is using generative AI to hawk games that don't existThe publisher is using the controversial technology to determine if players would want spinoffs for Call of Duty and Crash Bandicoot.Justin Carter, Contributing EditorMarch 3, 20252 Min ReadImage via Activision Blizzard.At a GlanceActivision Blizzard appears to be using generative AI to market games that don't exist.Activision Blizzard has been using generative AI to create online ads for games that do not actually exist.VGC spotted promos for three fake titlesGuitar Hero Mobile, Crash Bandicoot Brawl, and Call of Duty: Zombie Defenderon Instagram. The ads lead to mockup app store pages on a site called Geeklab, where upon clicking "Get," viewers are treated to the message: "Thanks for your interest! This isn't a real game, but could be some day! We'd love if you could answer this short survey, which could help inform the potential future of this game. Your feedback really matters to us!"Last week, the Call of Duty publisher confirmed players' suspicions that it was using the controversial technology to produce "some" in-game assets. At the moment, it's unclear what those specific assets are, and it only came out because of Valve's Steam conduct on studios disclosing if their game uses genAI in any capacity.Most use surrounding the technology has concerned its efficiency on game development in fields like art and voice acting. In an early 2024 survey, nearly 50 percent of developers said they were using the technology at work, with most of that use concerning finances or community and product management.Game publishers typically gauge interest for future game projects with surveys, like EA asking players if there was any appeal in remaking the Dead Space sequels. Activision Blizzard's new "advertising" is ethically dicey, both because it's one of the industry's biggest publishers, and because the ads are widely promoting titles which aren't real now and may possibly not become real in the future.Game Developer has reached out to Activision Blizzard about the ads, and will update when a comment is given.Read more about:Activision BlizzardXboxAbout the AuthorJustin CarterContributing Editor, GameDeveloper.comA Kansas City, MO native, Justin Carter has written for numerous sites including IGN, Polygon, and SyFy Wire. In addition to Game Developer, his writing can be found at io9 over on Gizmodo. Don't ask him about how much gum he's had, because the answer will be more than he's willing to admit.See more from Justin CarterDaily news, dev blogs, and stories from Game Developer straight to your inboxStay UpdatedYou May Also Like

Taiwan Semiconductor Manufacturing Co. the worlds biggest chipmaker will invest at least $100 billion to expand chip manufacturing in the US. During a press conference on Monday, President Donald Trump said the funding would go toward building two additional chip manufacturing facilities in Phoenix, Arizona.The $100 billion investment builds upon the $65 billion TSMC has already committed to building three Arizona factories, as well as the $6.6 billion the Biden administration awarded to TSMC under the CHIPS Act. TSMC began producing 4-nanometer chips at its Arizona plant in January, but its future factories are expected to make chips using 2nm or even more advanced process technology by the end of the decade, according to the companys website.Last year, TSMC pushed back the timeline for its second Arizona plant, saying it will open in 2027 or 2028 instead of 2026.We are producing the most advanced chip made on US soil with the success of our first plant, TSMC CEO C.C. Wei said during the press conference. Were going to create thousands of high-paying jobs and produce many AI chips. A recent report from The New York Times suggested that the Trump administration encouraged TSMC to take over Intels chipmaking factories.Last week, Apple similarly revealed its plans to invest over $500 billion in the US over the next four years, during which it plans to hire 20,000 employees and construct a server factory in Texas. The timing isnt a coincidence, as President Trump has said he will introduce tariffs on semiconductors and other goods as soon as April. On Tuesday, the US is also set to start imposing a 25 percent tariff on products from Mexico and Canada, along with an additional 10 percent tax on imports from China.

Dec 11, 2024Andrew J. HawkinsThe end of Cruise is the beginning of a risky new phase for autonomous vehiclesCath Virginia / The Verge | Photo from Getty ImagesEight years and $10 billion later, GM has decided to pull the plug on its grand robotaxi experiment.The automakers CEO, Mary Barra, made the surprise announcement late on Tuesday, arguing that a shared autonomous mobility service was never really in its core business. It was too expensive and had too many regulatory hurdles to overcome to make it a viable revenue stream. Instead, GM would pivot to privately owned driverless cars because, after all, thats what the people really wanted.Read Article >Dec 10, 2024Andrew J. HawkinsCruises robotaxi service will shut down as GM pulls its fundingPhoto by Smith Collection/Gado/Getty ImagesGeneral Motors said it would no longer fund its Cruise robotaxi service as it seeks to focus its spending on autonomous vehicle development specifically for personally owned vehicles. Now Cruise employees will be combined with GMs internal teams working on advanced driver assist systems, like Super Cruise, as well as its project to develop autonomous vehicles to sell to customers for personal use.Ultimately, the project became too expensive for GM to justify the huge amounts of money spent to prop it up. And the automaker found it increasingly difficult to convince its shareholders that the money-losing operation would eventually pay off. The robotaxi subsidiary lost a staggering $3.48 billion in 2023 and has been seen by some as an albatross for the automaker, sucking up cash and lacking a clear path to profits.Read Article >Aug 13, 2024Richard LawlerandWes DavisWaymo released a software update to solve its San Francisco honkfest.Waymo spokesperson Chris Bonelli explains why the problem started and why its autonomous cars are quieter now (shown in the livestream below):We recently introduced a useful feature to help avoid low speed collisions by honking if other cars get too close while reversing toward us. It has been working great in the city, but we didnt quite anticipate it would happen so often in our own parking lots. Weve updated the software, so our electric vehicles should keep the noise down for our neighbors moving forward.Aug 11, 2024Wes DavisA nightly Waymo robotaxi parking lot honkfest is waking San Francisco neighborsA Waymo car out on the job. Photo: Smith Collection / Gado / Getty ImagesIf youve ever wondered what happens to all those self-driving taxis when the world is asleep, one YouTube channel has you covered. Since the beginning of the month, software engineer Sophia Tung has been livestreaming a San Francisco parking lot that Waymo is renting to give its robotaxis somewhere to go during their downtime.Tung told The Verge via email that the company appeared to partially take over the lot on July 28th, then later took over the entire lot. Waymo recently opened up its robotaxi service to anyone in San Francisco.Read Article >Jun 7, 2024Sean HollisterAmazon will begin testing robotaxis in Austin and Miami too.Thats on top of San Francisco, Las Vegas, Seattle, and Zooxs original Foster City test. Note, though:We wont be deploying our purpose-built robotaxi or offering public rides in Austin or Miami just yet. Our target cities for our first commercial markets remain Las Vegas and San Francisco. But were exploring several cities for future commercial offerings after our initial launch.Were expanding testing to Austin and Miami[Zoox]May 14, 2024Andrew J. HawkinsWaymos robotaxis are under investigation for crashes and traffic law violationsPhoto by Justin Sullivan / Getty ImagesFederal safety investigators have opened a preliminary investigation into dozens of incidents involving Waymos driverless vehicles, including several single-party crashes and possible traffic law violations.The National Highway Traffic Safety Administrations Office of Defects Investigation is looking into 22 incidents in which Waymos robotaxis were the sole vehicle operated during a collision or exhibited driving behavior that potentially violated traffic safety laws.Read Article >Apr 27, 2024Wes DavisIn the first Autonomous Racing League race, the struggle was realThe first race of the Abu Dhabi Autonomous Racing League (A2RL) took place on the Yas Marina Abu Dhabi Grand Prix Formula 1 track today, and Im pleased to report that a race both began and ended. But the event was not without strife far from it. During qualifying time trials, the driverless Dallara Super Formula racers outfitted with cameras and software seemed to struggle mightily to complete a full lap.During the trials, cars randomly juked:Read Article >Apr 18, 2024Andrew J. HawkinsOrange cones are robot kryptonite.At least seven driverless Waymo vehicles blocked the on-ramp to 101 after being flummoxed by a construction site on Portrero Drive, according to video of the incident posted on Reddit. Waymo told TechCrunch the detour necessitated driving on the freeway, for which the vehicles were not approved without a human driver. To me, this is more evidence of the innate, robot-fighting powers of the humble orange cone.Apr 16, 2024Andrew J. HawkinsPeople are afraid of self-driving cars can the industry change that?Illustration by Alex Castro / The VergeSelf-driving cars have an image problem. They stumble into construction zones, block ambulances, run red lights, and even injure the occasional bicyclist or pedestrian.Forget the fact that most autonomous vehicles operate each day safely, anonymously, and without fanfare. There are hundreds in operation today in California, Arizona, Texas, and elsewhere, and the numbers are only going to increase. But when they do make mistakes, people tend to notice. Numerous public opinion polls have shown declining support for autonomous vehicles (AV) over the years and a rise in outright hostility toward the technology.Read Article >Apr 3, 2024Allison JohnsonUber Eats is now delivering burritos via Waymo in Phoenix.Waymo has been offering autonomous rides in Phoenix since late 2022, but now its providing transportation for a new kind of customer: your dinner. Starting today, Waymo will start making food deliveries for some Uber Eats customers.Uber works with a few other companies for autonomous food deliveries in other locations. For now, this program is limited to Tempe, Mesa, and Chandler, and only select merchants are included. Taco bout a slow rollout.Uber Newsroom[www.uber.com]Mar 14, 2024Andrew J. HawkinsThe Zoox come out at night.Amazons robotaxi company is expanding its operating conditions in the two cities where it has been testing its autonomous vehicles, Foster City and Las Vegas. The robotaxis will start driving at night, as well as in light rain and damp road conditions. They will also starting driving at speeds of up to 45 mph on multi-lane roads. And in Las Vegas, it will start tackling roads along the south end of the Strip. Zooxs purpose-built AVs (no steering wheel, no pedals) will start accepting real passengers later this year.Gadzooks! Its a Zoox at night! Image: ZooxMar 13, 2024Andrew J. HawkinsWaymo opens up in LA.The robotaxi company is ready to start inviting regular people on its waitlist (50,000 and counting) to use its fully driverless vehicles. The vehicles will only operate in a 63 square-mile section that includes Santa Monica and DTLA. And while the initial rides will be free, future rides will not thanks to a recent thumbs-up from regulators.The company plans to follow similar rollout in Austin, Texas, later this year. The future of autonomous vehicles still seems super cloudy, but Waymo is trying its best to prove the doubters wrong.Feb 12, 2024Andrew J. HawkinsCruise finally has a chief safety officer.Steve Kenner has held safety-related positions at Apple and Uber, as well as autonomous vehicle companies like Aurora, Kodiak Robotics, and Locomation. Hell be in charge of making sure Cruise adheres to safety standards as it seeks to re-deploy its robotaxis in San Francisco and beyond. The GM-owned company is trying to rebuild its reputation and get back its operations permit in the wake of an incident in which a pedestrian was stuck and dragged by one of its autonomous vehicles.Cruise Appoints Chief Safety Officer[getcruise.com]Jan 25, 2024Andrew J. HawkinsCruise wasnt hiding the pedestrian-dragging video from regulators it just had bad internetPhoto by Tayfun Coskun / Anadolu via Getty ImagesCruise, the self-driving car subsidiary of General Motors, tried to send a 45-second video to regulators of an incident in which one of its driverless cars dragged a pedestrian 20 feet but was hampered by internet connectivity issues, according to a report compiled by a law firm investigating the incident.The law firm, Quinn Emanuel Urquhart & Sullivan, was hired by Cruise to determine whether its executives misled regulators in the aftermath of the October 2nd incident in which a hit-and-run driver struck a pedestrian, knocking her into the path of a driverless Cruise vehicle. Its conclusions were detailed in a nearly 200-page report that was released today.Read Article >Nov 20, 2023Andrew J. HawkinsGMs big bet on driverless cars turns sourThen Cruise CEO Kyle Vogt introducing the companys driverless Origin shuttle. Photo By Carlos Avila Gonzalez/The San Francisco Chronicle via Getty ImagesTwo years ago, General Motors presented a vision for the future that involved zero crashes, zero traffic, and zero emissions. Today, that future seems further away than ever.The automakers driverless car subsidiary, Cruise, announced last night the resignation of Kyle Vogt as CEO. The decision came over a month after an incident in which a hit-and-run victim became pinned under a Cruise vehicle and then was dragged 20 feet to the side of the road. As a result, California Department of Motor Vehicles suspended Cruises permit to operate driverless cars in the state.Read Article >Oct 26, 2023Andrew J. HawkinsTFW your Uber driver is an empty seat.Starting today, Phoenix residents can use the Uber app to hail a ride in a driverless Waymo vehicle. The two companies former rivals turned frenemies (?) first announced the partnership earlier this year. Tellingly, its only available in Arizona, and not California, where tensions around robotaxis are starting to get, well, tense.Image: WaymoOct 13, 2023Andrew J. HawkinsHow will driverless cars talk to pedestrians? Waymo has a few ideasImage: WaymoFor years, developers have been working on ways for driverless cars to communicate intent to other road users, either through audio recordings or visual cues. Today, Waymo says it wants to be one of the first companies to put some of these methods into practice.The Alphabet-owned companys driverless Jaguar I-Pace vehicles will use their roof domes, which are wrapped in LED displays, to communicate messages to other road users. For now, the company is going with just two messages: for pedestrians in front of the vehicle, shifting grey and white rectangles meant to communicate that the vehicle is yielding to them, and for drivers behind the vehicle, a yellow pedestrian symbol to let them know theres a pedestrian crossing.Read Article >Oct 12, 2023Andrew J. HawkinsRobotaxis in the Bayou City.Houston is the next city in the US to get a robotaxi service, courtesy of Cruise, which just announced the launch today. The driverless vehicles will be available seven days a week from 9PM-6AM in Downtown, Midtown, East Downtown, Montrose, Hyde Park, and River Oaks neighborhoods. Robotaxi companies have been targeting bigger, more populous markets, as the pressure to start bringing in more revenue continues to grow. Waymo just started testing the waters in LA, and now Cruise is going after the fourth biggest city in the US.Cruises robotaxis will only operate at night to start out. Image: CruiseOct 11, 2023Andrew J. HawkinsWhat about drivers ed for driverless cars?This opinion piece in the New York Times argues were driving blind when it comes to autonomous vehicles, citing recent robotaxi crashes in San Francisco and the growing number of fatal Tesla Autopilot incidents. The writer argues that while the federal government regulates hardware, and the states oversee drivers, theres no one testing to see whether the software operating these vehicles is up to snuff. And that amounts to a loophole large enough for Elon Musk, General Motors and Waymo to drive thousands of cars through.Hmmm, where have I heard this argument before?Opinion | Autonomous Vehicles Are Driving Blind[The New York Times]Oct 10, 2023Andrew J. HawkinsWaymos robotaxis are now available to tens of thousands of people across all of San FranciscoPhoto by AFP / AFP TV / AFP via Getty ImagesWaymos robotaxi service area in San Francisco is expanding to 47 square miles of the city a significant expansion of the companys driverless ridehail operations.Last August, Waymo and its driverless rivals won approval from California regulators to operate their commercial service 24/7. The vote was a significant win for the tech industry, which has been battling criticism that its robot cars occasionally obstruct emergency vehicles and cause traffic jams.Read Article >Sep 23, 2023Wes DavisCalifornia governor vetoes a bill requiring humans in autonomous big rigsDriverless trucks from Waymo. Credit: WaymoGovernor Gavin Newsom has vetoed Assembly Bill 316, which would have required human attendants in driverless vehicles over 10,000 pounds, reports Reuters. The bill saw broad support among state legislators and was backed by the Teamsters and other labor organizations. At the moment,The governor wrote in his veto message that the bill is unnecessary for the regulation and oversight of heavy-duty autonomous vehicle technology, adding that the existing regulatory framework is sufficient.Read Article >Sep 14, 2023Andrew J. HawkinsCruise unveils a wheelchair-accessible robotaxi, with plans to launch in 2024Image: Cruise / Raymond Rudolph PhotographyCruise, the autonomous vehicle company backed by General Motors, revealed a wheelchair-accessible robotaxi that it says could start picking up disabled passengers as soon as next year.The reveal of the newly accessible robotaxi is a major step toward fulfilling the dream of people with vision, hearing, and mobility impairments, who have long held out hope that autonomous vehicles represent a new way of getting around.Read Article >Aug 19, 2023Wes DavisDriverless buses are tooling around San Franciscos Treasure Island in a test for the next few months.The autonomous shuttles offer free rides along a fixed, seven-stop route in the artificial islands center from August 2023 to April 2024 (via AP News).Called The Loop, the shuttle has no steering wheel, but an onboard attendant can take over with a handheld remote if needed, according to Insider. Its maker, Beep, previously tested it as a medical supply transport in Florida in 2020.Treasure Island Autonomous Shuttle Launches Public Service | San Francisco[sf.gov]Aug 17, 2023Andrew J. HawkinsWe knew Tesla didnt fix a flaw in Autopilot, and now we have their engineers on record admitting it.When two people die in very similar crashes years apart, the reason seems obvious. Autopilot, Teslas driver assist system, cant recognize trucks crossing the road. They knew it couldnt, and they didnt fix it. And now we have testimony from their engineers admitting this.Despite the companys knowledge that theres cross traffic or potential for cross traffic, the Autopilot at the time was not designed to detect that, according to testimony given in 2021 by company engineer Chris Payne that was excerpted in a recent court filing. Engineer Nicklas Gustafsson provided a similar account in a 2021 deposition.The family for one of the dead Tesla owners is seeking punitive damages in a lawsuit set to go to trial this October.Tesla Failed to Fix Autopilot After Fatal Florida Crash, Engineers Say[Bloomberg]Aug 15, 2023Andrew J. HawkinsRobotaxis are driving on thin iceIllustration by Alex Castro / The VergeThe day after California regulators handed driverless car companies a major victory, allowing them to expand their services without restriction in San Francisco, a herd of robotaxis decided to celebrate by breaking down in the middle of a busy street.According to several local news reports, 10 Cruise vehicles sat paralyzed in a busy intersection near the Outside Lands Music Festival, causing a traffic jam and drawing exasperation from witnesses. The company told KPIX that the music festival caused wireless connectivity issues with its vehicles. In other words, festivalgoers were overwhelming the cellular networks, making it difficult for Cruises vehicles to send and receive information.Read Article >

CrewAI is an open-source framework for orchestrating autonomous AI agents in a team. It allows you to create an AI crew where each agent has a specific role and goal and works together to accomplish complex tasks. In a CrewAI system, multiple agents can collaborate, share information, and coordinate their actions toward a common objective. This makes it possible to break down a problem into sub-tasks and have specialized agents tackle each part, much like a team of humans with different expertise.In this tutorial, well demonstrate a use case of multiple AI agents working together using CrewAI. Our example scenario will involve summarizing an article using three agents with distinct roles:Research Assistant Agent Reads the article and extracts the key points or facts.Summarizer Agent Takes the key points and concisely summarizes the article.Writer Agent Reviews the summary and formats it into a structured final output (for example, adding a title or conclusion).This collaborative workflow mimics how a team might work: one member gathers information, another condenses it, and a third polishes the presentation. We will implement this workflow with CrewAIs abstractions (Agents, Tasks, and Crew).Code Implementation!pip install crewai crewai-tools transformersFirst we install all the necessary packages for the project in one go. The crewai and crewai-tools packages provide the framework and additional utilities for orchestrating AI agents, while the transformers package from Hugging Face supplies pre-trained models for text processing tasks like summarization.from crewai import Agent, Task, Crew, ProcessHere we import the core classes from the CrewAI framework. The Agent class lets you define an AI agent with a specific role and behavior, Task represents a unit of work assigned to an agent, Crew orchestrates the collaboration among these agents, and Process sets the execution workflow (like sequential or parallel).# Define the agents' roles, goals, and backstories to satisfy the model requirements.research_agent = Agent( role="Research Assistant", goal="Extract the main points and important facts from the article.", backstory="You are a meticulous research assistant who carefully reads texts and extracts key points.")summarizer_agent = Agent( role="Summarizer", goal="Summarize the key points into a concise paragraph.", backstory="You are an expert summarizer who can condense information into a clear and brief summary.")writer_agent = Agent( role="Writer", goal="Organize the summary into a final report with a title and conclusion.", backstory="You are a creative writer with an eye for structure and clarity, ensuring the final output is polished.")We define three specialized AI agents using the CrewAI framework through the above code. Each agent is configured with a specific role, goal, and backstory, which instructs them on how to contribute to the overall task: the Research Assistant extracts key points from an article, the Summarizer condenses those points into a concise paragraph, and the Writer formats the final output into a polished report.# Example: Create tasks for each agentresearch_task = Task( description="Read the article and identify the main points and important facts.", expected_output="A list of bullet points summarizing the key information from the article.", agent=research_agent)summarization_task = Task( description="Take the above bullet points and summarize them into a concise paragraph that captures the article's essence.", expected_output="A brief paragraph summarizing the article.", agent=summarizer_agent)writing_task = Task( description="Review the summary paragraph and format it with a clear title and a concluding sentence.", expected_output="A structured summary of the article with a title and conclusion.", agent=writer_agent)The above three task definitions assign specific responsibilities to the respective agents. The research_task instructs the Research Assistant to extract key points from the article, the summarization_task directs the Summarizer to turn those points into a concise paragraph, and the writing_task asks the Writer to format the summary into a structured final report with a title and conclusion.# Create a crew with a sequential processcrew = Crew( agents=[research_agent, summarizer_agent, writer_agent], tasks=[research_task, summarization_task, writing_task], process=Process.sequential, verbose=True)print("Agents defined successfully!")Now, we create a Crew object that orchestrates the collaboration between the three defined agents and their corresponding tasks in a sequential workflow. With Process.sequential, each task is executed one after the other, and setting verbose=True enables detailed logging of the process.# Sample article text (as a multi-paragraph string)article_text = """Artificial intelligence (AI) has made significant inroads in various sectors, transforming how we work and live.One of the most notable impacts of AI is seen in healthcare, where machine learning algorithms assist doctors in diagnosing diseases faster and more accurately.In the automotive industry, AI powers self-driving cars, analyzing traffic patterns in real-time to ensure passenger safety.This technology also plays a crucial role in finance, with AI-driven algorithms detecting fraudulent transactions and enabling personalized banking services.Education is another field being revolutionized by AI. Intelligent tutoring systems and personalized learning platforms adapt to individual student needs, making education more accessible and effective.Despite these advancements, AI adoption also raises important questions.Concerns about job displacement, ethical use of AI, and ensuring data privacy are at the forefront of public discourse as AI continues to evolve.Overall, AI's growing influence across industries showcases its potential to tackle complex problems, but it also underscores the need for thoughtful integration to address the challenges that accompany these innovations."""print("Article length (characters):", len(article_text))In this part, we define a multi-paragraph string named article_text that simulates an article discussing AIs impact across various industries, such as healthcare, automotive, finance, and education. It then prints the character length of the article, helping to verify that the text has been loaded correctly for further processing.import redef extract_key_points(text): paragraphs = [p.strip() for p in text.split("\n\n") if p.strip()] key_points = [] for para in paragraphs: sentences = re.split(r'(?<=\.)\s+', para.strip()) if not sentences: continue main_sentence = max(sentences, key=len) point = main_sentence.strip() if point and point[-1] not in ".!?": point += "." key_points.append(point) return key_points# Use the function on the article_textkey_points = extract_key_points(article_text)print("Research Assistant Agent - Key Points:\n")for i, point in enumerate(key_points, 1): print(f"{i}. {point}")Here we define a function extract_key_points that processes an articles text by splitting it into paragraphs and further into sentences. For each paragraph, it selects the longest sentence as the key point (as a heuristic) and ensures it ends with proper punctuation. Finally, it prints each key point as a numbered list, simulating the Research Assistant agents output.from transformers import pipeline# Initialize a summarization pipelinesummarizer_pipeline = pipeline("summarization", model="sshleifer/distilbart-cnn-12-6")# Prepare the input for summarization by joining the key pointsinput_text = " ".join(key_points)summary_output = summarizer_pipeline(input_text, max_length=100, min_length=30, do_sample=False)summary_text = summary_output[0]['summary_text']print("Summarizer Agent - Summary:\n")print(summary_text)We initialize a Hugging Face summarization pipeline using the sshleifer/distilbart-cnn-12-6 model. It then joins the extracted key points into a single string and feeds it to the pipeline, which generates a concise summary. Finally, it prints the resulting summary, simulating the output from the Summarizer Agent.# Writer agent formattingtitle = "AI's Impact Across Industries: A Summary"conclusion = "In conclusion, while AI offers tremendous benefits across sectors, addressing its challenges is crucial for its responsible adoption."# Combine title, summary, and conclusionfinal_report = f"# {title}\n\n{summary_text}\n\n{conclusion}"print("Writer Agent - Final Output:\n")print(final_report)Finally, we simulate the Writer Agents work by formatting the final output. It defines a title and a conclusion and then combines these with the previously generated summary (stored in summary_text) using an f-string. The final report is formatted in Markdown, with the title as a header, followed by the summary and concluding sentence. It is then printed to display the complete, structured summary.In conclusion, this tutorial showed how to create a team of AI agents using CrewAI and have them work together on a task. We structured a problem (summarizing an article) into sub-tasks handled by specialized agents and demonstrated the end-to-end flow with example outputs. CrewAI provides a flexible framework to manage these multi-agent collaborations, while we, as users, define the roles and processes that guide the agents teamwork.Here is the Colab Notebook for the above project. Also,dont forget to follow us onTwitterand join ourTelegram ChannelandLinkedIn Group. Dont Forget to join our80k+ ML SubReddit. Asif RazzaqWebsite| + postsBioAsif 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.Asif Razzaqhttps://www.marktechpost.com/author/6flvq/DeepSeek AI Releases Smallpond: A Lightweight Data Processing Framework Built on DuckDB and 3FSAsif Razzaqhttps://www.marktechpost.com/author/6flvq/Researchers from UCLA, UC Merced and Adobe propose METAL: A Multi-Agent Framework that Divides the Task of Chart Generation into the Iterative Collaboration among Specialized AgentsAsif Razzaqhttps://www.marktechpost.com/author/6flvq/DeepSeeks Latest Inference Release: A Transparent Open-Source Mirage?Asif Razzaqhttps://www.marktechpost.com/author/6flvq/A-MEM: A Novel Agentic Memory System for LLM Agents that Enables Dynamic Memory Structuring without Relying on Static, Predetermined Memory Operations Recommended Open-Source AI Platform: IntellAgent is a An Open-Source Multi-Agent Framework to Evaluate Complex Conversational AI System' (Promoted)

Built a Computer Vision-Powered App Using Gemini in Under 15 Minutes No Training Required 0 like March 3, 2025Share this postAuthor(s): Areeb Adnan Khan Originally published on Towards AI. This member-only story is on us. Upgrade to access all of Medium.Machine Learning Algorithm Illustration: Source Getty ImagesComputer Vision is booming, and with the rise of multi modal AI models, its easier than ever to leverage its power. From industrial security checks to image classification and identification, the applications are rapidly expanding.But if youre just getting started and feeling overwhelmed by the sheer amount of information out there dont worry! This guide will walk you through performing image classification with almost 98% accuracy using zero-shot learning no prior training required.Zero-shot learning (ZSL) is a machine learning technique that allows models to classify images without having seen examples during training. This is incredibly powerful because:It eliminates the need for large, labeled datasets.It enables models to generalize across categories.It saves time and computational resources.By leveraging Pre-trained AI models, we can classify images on the fly using text-based descriptions a game-changer for those who want fast, accurate results without deep AI expertise.To get started, you need to install the genai library [1]. This is a generative AI library designed for IPython environments, making it an excellent tool for both learning and interacting with AI models.Run the following command to install Read the full blog for free on Medium.Join thousands of data leaders on the AI newsletter. Join over 80,000 subscribers and keep up to date with the latest developments in AI. From research to projects and ideas. If you are building an AI startup, an AI-related product, or a service, we invite you to consider becoming asponsor. Published via Towards AITowards AI - Medium Share this post

Author(s): William Lindskog Originally published on Towards AI. One-Size-Fits-All AI is Dead: The Rise of Personalized Federated LearningImage from Flower LabsFederated Learning (FL) has emerged as a powerful paradigm for training machine learning models across distributed datasets without requiring raw data to be centralized. This approach has enabled advancements in privacy-preserving AI, particularly in sensitive domains such as healthcare [1], finance [2], and autonomous systems [3]. However, a critical limitation of standard FL remains: global models trained across diverse data sources may not perform optimally for all individual clients.This is where Personalized Federated Learning (PFL) comes into play.Unlike traditional FL, which assumes a single global model can generalize across all clients, PFL acknowledges and addresses client heterogeneity by allowing models to be personalized for each client while still leveraging shared knowledge. Through techniques such as meta-learning, multi-task learning, and clustered model aggregation, PFL enables clients to retain a model tailored to their unique data distribution while benefiting from the collective intelligence of the network. [4]Illustration of client drift in FedAvg for 2 clients with 2 local steps [4].The Need for Personalization in Federated LearningTraditional FL aggregates model updates from multiple clients to create a single global model [5]. However, in real-world scenarios, data heterogeneity is the norm rather than the exception. The above image illustrates the problem of using a basic approach like Federated Averaging (FedAvg) where the global optima is different from local optima. Consider the following:Healthcare AI: A federated model trained across multiple hospitals may not generalize well to a particular hospitals unique patient demographics and medical equipment settings. [6]Autonomous Vehicles: Self-driving cars operating in different cities face distinct road conditions, weather patterns, and traffic laws. [7]Personalized Recommendation Systems: Users have highly individualized preferences, making a one-size-fits-all model suboptimal for applications like personalized content recommendations and adaptive learning platforms. [8]Thus, if we get an average solution, that doesnt mean it is a one-size-fits all solution. Given these challenges, PFL enables models to balance global generalization with local specialization and offers a way to leverage collective intelligence while maintaining local relevance.Approaches to Personalized Federated LearningSeveral strategies have been proposed to integrate personalization into federated learning. Below, we highlight key methodologies in PFL:Local Fine-TuningOne of the simplest approaches to Personalized Federated Learning is fine-tuning the global model on local client data after federated training. This method allows each client to adjust its model while still benefiting from the aggregated knowledge of the global model. Example: A practical implementation of this approach is FlowerTune LLM, a framework for federated fine-tuning of large language models (LLMs). FlowerTune enables multiple organizations or devices to collaboratively train an LLM on decentralized text data while preserving data privacy. [9]For example, in a medical AI setting, hospitals can fine-tune a federated LLM on local patient notes to improve its domain-specific language understanding. While the global model captures broad knowledge, each hospital benefits from a personalized version that adapts to its unique dataset. This approach is particularly powerful for LLMs deployed on personal devices, where models can be fine-tuned on a users own writing style, vocabulary, and preferences, resulting in more accurate and personalized AI assistants while maintaining user privacy.Clustered Federated LearningIn standard FL, a single global model is trained across all clients, assuming that a unified model can generalize well across diverse data distributions. However, in reality, clients often belong to different subpopulations, leading to non-i.i.d. (non-independent and identically distributed) data distributions. Rather than training a single model for all clients, clustered FL groups clients based on similar data distributions and trains separate models per cluster. [10] Example: Imagine a federated learning system for smart wearables that monitors heart rate, sleep patterns, and activity levels. A one-size-fits-all model would struggle to accommodate the variations between young athletes, middle-aged office workers, and elderly individuals.With Clustered FL, the system can automatically group users based on physiological patterns and train separate models for each lifestyle cluster. This ensures more accurate health recommendations while still leveraging federated learnings privacy-preserving benefits.Personalized Model ArchitecturesSome PFL approaches allow clients to train different model architectures while still benefiting from shared representations. These methods, such as layer-wise personalization, freeze certain parts of the model while letting others adapt locally [11]. This has successfully been deployed in projects and frameworks like Flower [9] and can be utilized for any task that requires personalization whether it is in healthcare, automotive, etc. Example: A federated speech recognition model might share a common phoneme recognition layer while personalizing user-specific accents and speech patterns.Future Directions in Personalized Federated LearningDespite the significant progress in Personalized Federated Learning (PFL), several challenges remain that must be addressed to unlock its full potential. One of the primary concerns is scalability as FL expands to millions of clients, the ability to efficiently personalize models without excessive computational overhead becomes crucial.Additionally, there is a growing need for adaptive federated learning, where the degree of personalization is dynamically adjusted based on real-world feedback. Instead of applying a fixed level of personalization across all clients, more intelligent FL algorithms could assess individual data distributions and automatically fine-tune the model adaptation process. This approach would ensure that clients receive the right level of personalization while maintaining stability in global model updates.As federated learning becomes more prevalent, personalization will be a key enabler of its success across diverse applications. Research in this area continues to evolve, with frameworks like Flower providing scalable and flexible solutions for real-world federated learning deployments.Personalized Federated Learning represents a significant step forward in distributed AI. By allowing models to benefit from global knowledge while adapting to local needs, PFL unlocks new possibilities for privacy-preserving, scalable, and adaptive AI systems. As research progresses, we can expect even more sophisticated personalization techniques that bring FL closer to achieving its full potential across industries. For those working in AI, now is the time to explore PFL because one-size-fits-all AI is no longer enough.References[1] Antunes, Rodolfo Stoffel, et al. Federated learning for healthcare: Systematic review and architecture proposal. ACM Transactions on Intelligent Systems and Technology (TIST) 13.4 (2022): 123.[2] Wen, Jie, et al. A survey on federated learning: challenges and applications. International Journal of Machine Learning and Cybernetics 14.2 (2023): 513535.[3] Zeng, Tengchan, et al. Federated learning on the road autonomous controller design for connected and autonomous vehicles. IEEE Transactions on Wireless Communications 21.12 (2022): 1040710423.[4] Tan, Alysa Ziying, et al. Towards personalized federated learning. IEEE transactions on neural networks and learning systems 34.12 (2022): 95879603.[5] H. B. McMahan, E. Moore, D. Ramage, S. Hampson, and B. A. y Arcas, Communication-efficient learning of deep networks from decentralized data, in Int. Conf. on Artificial Intelligence and Statistics, 2017.[6] Lu, Zili, et al. Federated learning with non-iid data: A survey. IEEE Internet of Things Journal (2024).[7] Li, Beibei, et al. FEEL: Federated end-to-end learning with non-IID data for vehicular ad hoc networks. IEEE Transactions on Intelligent Transportation Systems 23.9 (2022): 1672816740.[8] Javeed, Danish, et al. Federated learning-based personalized recommendation systems: An overview on security and privacy challenges. IEEE Transactions on Consumer Electronics 70.1 (2023): 26182627.[9] Beutel, Daniel J., et al. Flower: A friendly federated learning research framework. arXiv preprint arXiv:2007.14390 (2020).[10] Ma, Jie, et al. On the convergence of clustered federated learning. arXiv preprint arXiv:2202.06187 (2022).[11] Arivazhagan, M., et al. (2019). Federated Learning with Personalization Layers. arXiv preprint arXiv:1912.00818.Join thousands of data leaders on the AI newsletter. Join over 80,000 subscribers and keep up to date with the latest developments in AI. From research to projects and ideas. If you are building an AI startup, an AI-related product, or a service, we invite you to consider becoming asponsor. Published via Towards AI

Its March, and that can only mean one single solitary thing. Thats right: new LEGO sets are available. And unlike last month, there are a ton of new sets that will definitely be of interest to the IGN audience. Were talking sets based on Star Wars, Jurassic World, Harry Potter, Marvel, and more. New LEGO BrickHeadz Transformers Sets Are AvailableLEGO BrickHeadz Optimus Prime Robot & Vehicle$19.99 at LEGO StoreI love the BrickHeadz series of LEGO sets. Theyre just adorable. Theyre kind of like pixelated NES-style renderings of favorite characters. These new sets feature the Transformers Optimus Prime and Bumblebee. These sets are two-in-one, letting you build the Autobots in both robot and vehicle forms to display next to each other. The vehicle versions have spinning wheels, so you can take them off the stand to play with them, which is pretty neat.LEGO Star Wars: AT-AT Driver Is AvailableLEGO Star Wars AT-AT Driver Helmet$69.99 at LEGO StoreAT-AT Driver Helmet - $69.99Theres no shortage of awesome helmets in the Star Wars franchise, and LEGO has been capitalizing on that fact for years now with awesome helmet sets. The latest is the very cool AT-AT Driver helmet, which looks like a standard Stormtrooper helmet but even cooler with its red details and whatever those circles on its chin are. Looks like another entry in a long line of the best LEGO Star Wars sets.LEGO Horizon Adventures: Aloy & Varl vs. Shell-Walker & Sawtooth Is OutOut March 1LEGO Horizon Adventures Aloy & Varl vs. Shell-Walker & Sawtooth$44.99 at AmazonAlloy & Varl vs. Shell-Walker & Sawtooth - $44.99Heres another set based on the PlayStation series Horizon. This one is from Horizon Adventures, which landed on all major platforms and is intended for a younger audience. So, fittingly, this LEGO set is geared toward the 9+ age group. It features the two main characters battling some iconic robo-dinosaurs.LEGO Art Vincent van Gogh - Sunflowers Is OutOut March 1LEGO Art Vincent van Gogh Sunflowers$199.99 at LEGO StoreVincent van Gogh - Sunflowers - $199.99Vincent van Gogh is one of the best and most famous painters in the history of the world. LEGO has already celebrated the artists most famous painting with LEGO Starry Night. But now theres more, thanks to a new set devoted to his remarkable Sunflowers paintings. Check out our We Build LEGO Vincent van Gogh - Sunflowers feature to read about the building process and to learn about a very cool art history Easter egg the designers built into the set.LEGO Harry Potter Knight Bus Adventure Is OutLEGO Harry Potter Knight Bus AdventureKnight Bus Adventure - $49.99If youre a fan of LEGO Harry Potter sets, or you just like super-sized triple-decker buses, you can pick up the LEGO Harry Potter Knight Bus for the reasonable price of $49.99. The memorable bus appeared in The Prisoner of Azkaban, and you can build the very purple vehicle now.LEGO Bluey Sets Are Up for PreorderOut June 1LEGO Bluey: Blueys Family House$69.99 at AmazonBluey's Family House - $69.99Blueys Beach & Family Car Trip - $29.99Playground Fun with Bluey and Chloe - $19.99Bluey's Family House with Memory Game (Duplo) - $69.99Ice Cream Trip with Bluey (Duplo) - $29.99It can be hard for parents to find TV shows to watch with their kids that dont make them (the parents) want to claw their eyes/ears into mangled appendages. Take it from a parent: Bluey is one of the good ones. And the Australian animated franchise has finally made its way to the world of building blocks in a handful of cool new sets. Preorder one (or more) and build it with your kids when they release on June 1.LEGO Jurassic World Dinosaur Fossils: Tyrannosaurus Rex Is Out March 15Out March 15LEGO Jurassic World Dinosaur Fossils: Tyrannosaurus rex$249.99 at LEGO StoreHeres one for the Spielberg/Crichton fans. The block-building company has announced a huge new LEGO set for adults based on the unkillable dino franchise. This ones based on the original (and best by far) film: LEGO Jurassic World Dinosaur Fossils: Tyrannosaurus Rex. The set is a T-rex skeleton that measures over three feet long once its built. It comes with minifigs of Dr. Ellie Sattler and Dr. Alan Grant. Well have a We Build feature ready to go at release. Speaking of, if youre a LEGO Insider (sign up here for free), you can order this set on March 12. Everyone else will have to wait until March 15.New LEGO Creator 3-in-1 Sets Are AvailableLEGO Creator 3-in-1 - Record Player with Flowers$29.99 at LEGO StoreTheres no better bang for your buck in LEGO land than the LEGO Creator 3-in-1 series. Each of these sets has all the pieces you need to build not one set, but three sets (using the same pieces, so you cant build all three at once). The record player with flowers set, for instance, lets you build a turntable, a radio, and an olde-timey microphone.New LEGO Disney Sets Are AvailableLEGO Disney Moana 2 Heihei$39.99 at AmazonLilo and Stitch Beach House - $89.99Maleficents and Cruella De Vils Dresses - $69.99Cinderella's Dress - $39.99Heihei - $39.99Some excellent new LEGO Disney sets are out in March. My personal favorite is Heihei, the vapid chicken from Moana 2. The set looks great, getting the birds big dumb lovable eyes just right. Also check out more of the best LEGO Disney sets.New LEGO Minecraft SetsLEGO Minecraft The Ghast Balloon Village Attack$69.99 at AmazonThe Ghast Balloon Village Attack - $69.99The Parrot Houses - $69.99Woodland Mansion Fighting Ring - $49.99The Trial Chamber - $39.99With the Minecraft Movie hitting theaters soon, its no surprise that LEGO is beefing up its (already huge) collection of LEGO Minecraft sets. Four new ones arrive this month.More New LEGO Sets for March 2025LEGO Architecture Trevi Fountain$159.99 at AmazonNinjago City Workshops - $249.99French Cafe - $79.99The Evolution of STEM - $79.99Fast & Furious Toyota Supra MK4 - $59.99Scaled-Up Racing Driver Minifigure - $54.99Plenty of other awesome looking sets are out this month, including a handful of new LEGO Ninjago sets, a bunch of LEGO car sets, and more.Chris Reed is a deals expert and commerce editor for IGN. You can follow him on Bluesky @chrislreed.com.