I felt like I opened a can of worms when I asked Samsung about its stance on AI photo editing at Galaxy Unpacked back in January. “There is no such thing as a real picture,” Patrick Chomet, Samsung’s Head of Customer Experience, told me at the time – a nihilism-tinged soundbite that, in fairness, came as part of a nuanced and perfectly valid philosophical commentary on the nature of photography (in the same interview, Chomet rightly talked up the importance of validating authenticity).
Marques Brownlee recently included Chomet’s quote in an X post highlighting the differing opinions of Samsung, Apple, and Google on this very subject. Still, ultimately, all of the major phone manufacturers are offering similar editing features (Magic Eraser, Clean Up, AI Eraser, and so on).
Qualcomm, for its part, acknowledges that AI photo editing is an exciting new frontier but also one fraught with awkward ethical implications. During a roundtable interview at Snapdragon Summit, the company’s Mobile Handset SVP, Chris Patrick, explained why he believes that AI photo editing “is not as simple as a real image and fake image.”
LOL “What is a photo?”Samsung: “Actually, there is no such thing as a real picture”Google: “authentic to your memory and to the greater context, but maybe isn’t authentic to a particular millisecond”Apple: “a personal celebration of something that really, actually happened” https://t.co/mwy3Yb9KaYSeptember 23, 2024
“It’s an interesting question, because the human brain is not digital,” the former engineer explained. “When you perceive the sunset outside, you don’t perceive the sunset absent of context. You know where you are. You know what you’re smelling. You know what you’re hearing, You know what you’re feeling. You know what a sunset looks like, right? All of that is part of how you perceive the space.”
Advertisement
“So then, when we capture an image, should it just be the raw response from the sensor? Or is it right to include context in how that picture is created? I’m not an expert. I don’t have a PhD in this topic, but I do think it’s not as simple as a real image and fake image. I think context matters, and extracting the very best we can from the whole situation is an accurate reflection of what the eye and the brain do as well.”
When you perceive the sunset outside, you don’t perceive the sunset absent of context.Chris Patrick
“[However],” Patrick continued, “if there’s a picture of Rui [Guo, Honor CMO, also in attendance at the roundtable] and I on the moon, it should be very clear that that is not, in fact, a picture of Rui and I on the moon. That’s not correct – that’s manipulated.
“So, for us, it’s very clear that there has to be some mechanism to distinguish the other extreme – those manipulated images – for people. We’re building incredible technology together but we want to make sure that, in the end, it does not have a destructive impact, where people can no longer tell the difference between truth and fiction.”
Sign up for breaking news, reviews, opinion, top tech deals, and more.
Advertisement
More AI is on the way
Honor CMO Ray Guo announcing the Honor Magic 7 Pro on stage at Snapdragon Summit (Image credit: Honor)
Honor CMO Rui Guo also attended the Snapdragon Summit roundtable, having just announced that the Honor Magic 7 Pro – slated for release in Europe next year – will launch with the industry’s first on-device generative AI portrait enhancement. So, is the company using AI to make portraits appear better than they actually are? Or is the aim to mimic real-life details that Honor’s hardware can’t capture on its own?
“On the one hand,” Guo explained, “we have to make sure that the photo itself is good by modern consumer standards. Everyone knows what [constitutes] a good and bad photo. But on the other hand, we definitely want to retain the authenticity of the photo. We’re not trying to Photoshop the photo for the consumer, right? We should give the authority to the users at end of the day.”
And that, I think, is the crux of the matter. Tools like Magic Eraser and Clean Up become ethically dubious when users – or, more worryingly, politicians – attempt to pass tampered images off as real, but as a means of occasionally removing strangers from family vacation photos, AI photo editing is, for many people, undeniably useful.
Neither Qualcomm nor Honor is forcing these tools onto their consumers – if you want to take a photo in the same way as you always have done, you still can. Rather, companies are giving users the freedom to choose how to capture moments that matter to them. Sure, it’s a can of worms, but as Qualcomm’s Chris Patrick noted, “it’s not as simple as real image and fake image.”
Arm has taken its feud with Qualcomm to the next level, two years after filing a lawsuit against its former close partner. According to Bloomberg, the British semiconductor company has canceled the architecture license allowing Qualcomm to use its intellectual property and standards for chip design. As the news organization notes, Qualcomm, like many other chipmakers, uses Arm’s computer code that chips need to run software, such as operating systems. Arm has reportedly sent Qualcomm a 60-day notice of cancelation — if they don’t get to an agreement by then, it could have a huge impact on both companies’ finances and on Qualcomm’s operations.
The SoftBank-backed chipmaker sued Qualcomm in 2022 after the latter purchased a company called Nuvia, which is one of its other licensees. Arm argued that the US company didn’t obtain the necessary permits to transfer Nuvia’s licenses. As such, Nuvia breached their contract and it had terminated its licenses, Arm explained in its lawsuit. Qualcomm has been using Nuvia-developed technology in the chips designed for AI PCs, such as those from Microsoft and HP. But Arm wants the company to stop using Nuvia-developed tech and to destroy any Arm-based technology developed prior to the acquisition.
Qualcomm will have to stop selling most of the chips that account for its $39 billion in revenue, Bloomberg says, if the companies don’t resolve the issue within the next 60 days. It seems the US chipmaker believes this is a tactic by Arm to threaten its business and to get higher royalties, because its spokesperson told Bloomberg and the Financial Times: “This is more of the same from Arm — more unfounded threats designed to strong-arm a longtime partner, interfere with our performance-leading CPUs, and increase royalty rates regardless of the broad rights under our architecture license.” Qualcomm also accused Arm of attempting to disrupt the legal process, called its grounds for licensing termination “completely baseless” and said that it’s confident its “rights under its agreement with Arm will be affirmed.”
Meanwhile, an Arm spokesperson told us: “Following Qualcomm’s repeated material breaches of Arm’s license agreement, Arm is left with no choice but to take formal action requiring Qualcomm to remedy its breach or face termination of the agreement. This is necessary to protect the unparalleled ecosystem that Arm and its highly valued partners have built over more than 30 years. Arm is fully prepared for the trial in December and remains confident that the Court will find in Arm’s favor.”
Advertisement
Update, October 23, 2024, 11:33PM ET: This story has been updated to add Arm’s statement.
Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn More
Model merging is a fundamental AI process that enables organizations to reuse and combine existing trained models to achieve specific goals.
There are various ways that enterprises can use model merging today, but many approaches are complex. A new approach known as Differentiable Adaptive Merging (DAM) could be the answer, providing a solution to the current challenges of model merging. DAM offers an innovative solution to combining AI models while potentially reducing computational costs.
Arcee AI, a company focusing on efficient, specialized small language models, is leading the charge on DAM research. The company, which raised funding in May 2024, has evolved from providing model training tools to becoming a full-fledged model delivery platform with both open-source and commercial offerings.
Advertisement
How DAM creates a new path forward for model merging
Merging can help companies combine models specialized in different areas to create a new model capable in both areas.
The basic concept of merging data is very well understood with structured data and databases. However, merging models is more abstract than merging structured data, as the internal representations of the models are not as interpretable.
Thomas Gauthier-Caron, research engineer at Arcee AI and one of the authors of the DAM research explained to VentureBeat that traditional model merging has often relied on evolutionary algorithms. That approach can potentially be slow and unpredictable. DAM takes a different approach by leveraging established machine learning (ML) optimization techniques.
Gauthier-Caron explained that DAM aims to solve the problem of complexity in the model merging process. The company’s existing library, MergeKit, is useful for merging different models, but it is complex due to the various methods and parameters involved.
Advertisement
“We were wondering, can we make this easier, can we get the machine to optimize this for us, instead of us being in the weeds tweaking all of these parameters?” Gauthier-Caron said.
Instead of just mixing the models directly, DAM adjusts based on how much each model contributes. DAM uses scaling coefficients for each column in the models’ weight matrices. It automatically learns the best settings for these coefficients by testing how well the combined model performs, comparing the output with the original models and then adjusting the coefficients to get better results.
According to the research, DAM performs competitively with or better than existing methods like evolutionary merging, DARE-TIES and Model Soups. The technology represents a significant departure from existing approaches, according to Gauthier-Caron. He described evolutionary merging as a slow process, where it’s not entirely clear up front how good the result will be or how long the merge process should run.
Merging is not an Mixture of Experts approach
Data scientists combine models in many different ways. Among the increasingly popular approaches is the Mixture of Experts (MoE).
Advertisement
Gauthier-Caron emphasized model merging with DAM is something very different from MoE. He explained that MoE is a specific architecture that can be used to train language models.
The basic concept behind model merging is that it starts from the point where the organization already has trained models. Training these models usually costs a lot of money, so engineers aim to reuse existing trained models.
Practical applications and benefits of DAM for enterprise AI
One of DAM’s key advantages is its ability to combine specialized models efficiently.
One such example provided by Gauthier-Caron is if an organization wanted to combine a Japanese model with a math model. The goal of that combination is to make a model that’s good at math in Japanese, without the need to retrain. That’s one area where DAM can potentially excel.
Advertisement
The technology is particularly relevant for enterprise adoption of generative AI, where efficiency and cost considerations are paramount. Helping to create more efficient ways of operating at reduced cost is a key goal for Arcee overall. That’s why DAM research is important to both the company and ultimately its users too.
“Enterprise adoption of gen AI boils down to efficiency, availability, scalability and cost,” Mark McQuade, co-founder and CEO of Arcee AI told VentureBeat.
VB Daily
Stay in the know! Get the latest news in your inbox daily
With TechCrunch Disrupt 2024 just days away, we’re gearing up for an incredible three-day event packed with interactive sessions! From October 28-30 at Moscone West in San Francisco, dive into Q&A conversations with a panel of industry experts around the most pressing issues and cutting-edge tech trends to help empower your growth.
Don’t miss your chance to save and engage in these first come, first served Breakouts! Register now to lock in up to $400 in savings on select tickets before door prices increase. Want to bring a friend? Take advantage of the Expo+ 2-for-1 Pass and bring a plus-one at half the cost of a single Expo+ Pass. These offers are valid through October 27.
Discover the complete roster of Breakout Sessions below, spread across two stages at Disrupt.
Breakout Session Agenda
The Future of High-Growth Tech: Beyond the Apple App Store
As the tech landscape evolves, the next wave of high-growth, high-impact companies may not emerge from the Apple App Store, but rather from the Microsoft app store. This shift signals a broader transformation in historically slow-moving industries like legal tech, where innovative solutions are challenging the status quo. Buyers in these sectors are increasingly open to embracing change, paving the way for a new era of technological advancement. Join our panel of experts as we explore how AI and other emerging technologies are driving this evolution, and what it means for the future of knowledge work in industries traditionally resistant to rapid change.
How AI Is Supercharging Tools for Knowledge Workers
Advancements in AI are enabling new tooling that will 10x the productivity of knowledge workers by reducing monotonous, repetitive tasks. These advances have also unlocked opportunities for more creativity and experimentation. This panel explores the latest in professional services tooling and explores how companies can maximize performance and productivity. We will also explore the future of tooling for knowledge workers and how emerging breakthroughs might be applied. Let’s explore the future of work.
Generative AI: Beyond the Hype — Building Real-World Applications
Join this interactive session to explore the practical applications of generative AI. We’ll break down the different types of generative models, discuss their strengths and limitations, and showcase inspiring use cases across various industries.
Beyond Snowflake and Databricks: Insights from the Frontlines of Data Transformation Disruption
By 2025, our global data volume will reach 175 zetabytes, a figure that is 50% more than 2023. But while the wealth of data grows, it remains unwieldy to use. Poor data quality costs organizations an average of $12.9 million annually. As organizations grapple with the exponential growth of data, they need better data transformation solutions to process it. The worldwide spending on such digital transformation solutions is forecasted to reach $3.9 trillion by 2027. In other words, the market is ripe for challenging the status quo, even despite the continued growth of data darlings Snowflake and Databricks. Tomasz can speak to the future of data in the context of SF and DB’s direct competition, and how startups are tackling these challenges head-on amidst the acceleration of two industry giants.
Join us for an exciting panel discussion featuring three leading female venture capitalists from Progressive Ventures and How Women Invest. This session will provide you with essential insights into the funding world, focusing on the latest trends, what investors look for, and effective strategies for women founders to stand out.
Powering Ahead: The Future of Energy & Infrastructure
Join us for “Powering Ahead: The Future of Energy & Infrastructure” featuring Troy Helming, a visionary leader in renewable energy and successful entrepreneur. In our breakout session, we’ll dive into the latest innovations shaping the energy sector, including advancements in renewable technologies, grid modernization, and sustainable infrastructure development. We’ll also share best practices for building a startup, offering practical advice drawn from Troy’s experiences. Attendees will gain insights into how these trends are transforming the way we produce, distribute, and consume energy, and how to capitalize on emerging opportunities. Don’t miss this opportunity to learn from an industry expert about the challenges and prospects that lie ahead in the quest for a cleaner, more efficient energy future.
IPO or Bust? Tactical Approaches for Late-Stage Success
With rising interest rates and stricter regulatory scrutiny on acquisitions, many growth and late-stage companies that once secured billions in funding are struggling to achieve a successful outcome. In this session, two experienced investors will debate the state of late-stage venture, what they’re seeing in terms of deal activity, if this market is actually coming back as significantly as some are reporting, and where they see the landscape headed in 2025. As part of this, they will dig into what’s going on with IPO markets, whether/when they will open back up, and how this is impacting exit strategies for founders.
The Future of Go-to-Market in the AI Era
Led by Jane Alexander, CapitalG; Chris Klayko, Databricks; Kareem Amin, Clay; Austin Hughes, Unify
From auto-generated outbound messages to AI-written blog posts, AI is fundamentally changing the way that companies go to market. Come discuss the tension between AI-powered automation and human creativity with the founders building these products and the leading practitioners who use them.
Secrets to Actually Being Good at Startup PR in 2024
The communications and PR landscape has dramatically shifted in recent years: new mediums (Substack, podcasts) have spiked in influence and popularity, fueling new strategies (“going direct”) for shaping and amplifying public perceptions — all while an increasingly critical media landscape continues to hold all types of industry stakeholders accountable. How can early and growing companies responsibly navigate these changes and be successful in building an inspiring brand? This roundtable will dissect the growing breadth and importance of different comms functions: the under-appreciated, consequential nature of internal comms, the increasingly indispensable need for owned channels, the pros and cons of political takes amidst an election year, and how to actually get meaningful press coverage for a company. The tech communications industry has arguably never been more dynamic.
Beyond the Wrapper: Building and Raising Organically with AI
Many startups today fundraise on the promise of being an “AI-first company.” But how do investors distinguish between companies that organically have AI at their core and those that are “GPT wrappers”? How can founders demonstrate that AI is intrinsic to their business and not an add-on intended to ride the wave of AI interest? This session addresses these and related questions by focusing on meaningfully incorporating AI into your startup and getting funded by investors. We use cutting-edge fundraising stats from DocSend to frame the conversation and ask founders and VCs how AI is changing fundraising expectations and how startups can organically incorporate AI into their businesses while staying ahead of increasingly sophisticated investor interest.
Smaller, Faster, Smarter: How Tiny AI Is Democratizing AI Technology Starting with the Smart Home Camera
Led by Roeland Nusselder and Tony Fadell, Plumerai
Advertisement
Nest founder, former SVP of Apple’s iPhone and iPod teams, and principal at Build Collective, Tony Fadell joins Plumerai’s co-founder and CEO Roeland Nusselder to discuss how Tiny AI is democratizing AI. By making AI smaller, efficient, and cost-effective, Tiny AI is paving the way for widespread adoption across industries, fostering innovation, and putting the power of AI into the hands of many. Tiny AI is shaping a future where intelligence is embedded all around us, enhancing our daily lives in ways we’ve only dreamed of. The future is here, and it’s smaller than you think!
In the Webby Awards’ annual thought leadership series focused on the trends and consumer insights shaping the internet, this year the leading award for internet excellence will explore how brands, marketers, digital creatives, and technologists are embracing chronically online culture to forge more creative and sustainable connections with audiences. The presentation, titled “It’s giving brainrot: How chronically online culture is taking us from the niche to the nonsensical—and why that can be a good thing,” will be led by the Webby Awards general manager Nick Borenstein and will feature insights and social listening data from Meltwater, along with trends from over 13,000 submissions to the 28th Annual Webby Awards. Following the presentation, Margaret Johnson (chief creative officer, Goodby, Silverstein & Partners), David Mogensen (VP of Global Marketing, Uber), and Monica Khan (co-founder, Bay Area Creator Economy) will join for a fireside chat to share their insights into this cultural moment.
Is Your AI Deployment an Advantage or an Embarrassment? Here’s How to Know
Rushed AI deployments can translate to embarrassing incidents, reputational damage, and financial loss. Brands like Adobe, Snap, and Anthropic have joined a growing list of companies embracing AI red teaming to deploy AI responsibly and find emerging threats before bad actors. HackerOne will share tales from the front lines of AI safety and security, so you know how to avoid AI embarrassments — from the circumvention of AI guardrails to harmful content generation. You’ll learn how top companies use AI red teaming and actionable ways to reduce AI risk that extend your AI advantage.
Scaling Technical Startups: Navigating Growth, Positioning, and Competitive Pressure
It’s a challenging time to scale a technical or data-focused startup — competition for customers, fundraising, and top talent continues to increase. This session will break down unique, timely challenges for technical founders and teams, and provide actionable advice to thrive in this highly competitive market. Attendees will take away best practices in product development, open source vs. closed source strategies, and how to position and message effectively —especially in pre-revenue stages for investor appeal. Technical founders will learn how to navigate the complexities of fundraising, with a focus on raising Series A for data-driven businesses, and why scaling technical or infrastructure companies requires a different playbook compared to traditional SaaS startups.
Deep Tech in Winter: How to Win Investors in 2024
Led by Po Bronson, Pae Wu, and Duncan Turner, SOSV
Advertisement
Venture’s deep chill has been extra frigid for deep tech startups due to their reputation for long timelines and big capital needs. But investors are still game for deep tech startups that are smart about balancing product scope and time to market. Vertical integration and industrial scale-ups are out; selling innovation into existing supply chains, now that’s smart. SOSV general partners Duncan Turner, Dr. Pae Wu, and Po Bronson help oversee the launch of about 75 deep tech startups a year and work with hundreds of deep tech co-investors. They will discuss what’s getting investors to “yes” to deep tech investments now, and take questions from the audience.
Bringing the Outside In: Connecting Startups with Large Banks to Power the Future of Finance
Arvind, along with Citi partners and portfolio company Quantifind, will discuss Citi Ventures’ approach to working with startups and how governance and risk management are essential to responsible innovation. He will also discuss how attendees can peer through their own crystal ball to predict the next big thing in tech and finance, from AI to the fintech revolution and beyond.
The Age of Technical Engineering Founders: How They Are Driving AI Innovation
In the evolving AI landscape, innovation is being powered by solutions to deeply technical problems that require leaders to take on a much more hands-on, technical approach. We have moved away from the business founder/CEOs of the past and into a new age where engineering founders are increasingly more common. This panel, composed of leaders from Honeycomb, MinIO, Atlan, and Yugabyte will discuss why engineering skills are critical for the modern leader’s role. They will share specific examples of important skill sets, and how founders can position themselves for long-term success as technology leaders in the future.
Founder Mode: AI Startups in Learning, Health, and Autonomous Agents
Learn how today’s top founders are using AI to solve real-world problems. The panel experts will include MultiOn, which creates AI agents on the web to solve a variety of problems such as booking flights, shopping, and internet research; Miri.ai, which offers health and wellness AI coaches; and LearnTube, which uses generative AI to create instant learning courses, curriculum, quizzes, and certifications. Panel discussion and Q&A on how to launch, build, grow, and fund an AI startup.
How to Stand Out Amongst the AI Wave: Strategies for Success in Enterprise Sales
Companies worldwide are actively investing in AI deployments across a wide range of use cases, and thousands of startups have emerged to fill these needs. This breakout session features Rudina Seseri, founder and managing partner of Glasswing Ventures, and Marc Boroditsky, president of Revenue at Cloudflare. They explore how AI has created a new paradigm shift in selling to enterprises, what the largest companies are using as criteria when considering their purchases, and how to avoid false indications of interest. Attendees will learn how to optimize scarce time and resources to build a truly valuable and viable product.
Building AI Agents — for Product Leaders & Founders
This is a live, hands-on workshop tailored for product leaders, aimed at providing a practical introduction to agentic products. You’ll gain a clear understanding of what AI agents are, how they function, and where they fit into product strategies. The workshop will guide you step by step through the process of building two AI agents from scratch, giving you a strong grasp of how to apply these tools to solve business problems, automate processes, and enhance user experiences. Bring your laptops to this interactive workshop!
Safety, Trust, and Profit: Anticipating Misuse to Build Safer Products and Attract Investment
In today’s startup world, one data breach or safety failure can destroy your reputation and growth. For founders, the challenge is how to scale while ensuring user safety, privacy, and security, all within budget. VCs are focusing on startups with strong ESG practices, prioritizing those that address safety early. Failing to comply with data laws, such as COPPA, can result in penalties up to $170 million.
This session covers how to build safeguards to prevent tech misuse, like Apple AirTag’s misuse for stalking. With one in three women globally experiencing violence and 32 million child exploitation reports submitted to NCMEC in one year, tech’s role in abuse is a rising threat. We will also explore how open source security tools can cut costs by 55% and how addressing safety and security early prevents legal risks and attracts ethical investors, and most importantly protects the vulnerable.
Why are stablecoins beginning to take off as a payments product? What are the most interesting examples and use cases? What technology will they replace — and where do they go from here?
Advertisement
Don’t miss these insightful Breakout Sessions
The only way to join these first come, first served Breakout Sessions is by registering for Disrupt 2024. Any pass grants you full access to these sessions. Register today and save up to $400 before prices rise at the door or get the Expo+ 2-for-1 Pass. Lock in your discounted ticket here.
A lawsuit has been filed against Character.AI, its founders Noam Shazeer and Daniel De Freitas, and Google in the wake of a teenager’s death, alleging wrongful death, negligence, deceptive trade practices, and product liability. Filed by the teen’s mother, Megan Garcia, it claims the platform for custom AI chatbots was “unreasonably dangerous” and lacked safety guardrails while being marketed to children.
As outlined in the lawsuit, 14-year-old Sewell Setzer III began using Character.AI last year, interacting with chatbots modeled after characters from The Game of Thrones, including Daenerys Targaryen. Setzer, who chatted with the bots continuously in the months before his death, died by suicide on February 28th, 2024, “seconds” after his last interaction with the bot.
Accusations include the site “anthropomorphizing” AI characters and that the platform’s chatbots offer “psychotherapy without a license.” Character.AI houses mental health-focused chatbots like “Therapist” and “Are You Feeling Lonely,” which Setzer interacted with.
Garcia’s lawyers quote Shazeer saying in an interview that he and De Freitas left Google to start his own company because “there’s just too much brand risk in large companies to ever launch anything fun” and that he wanted to “maximally accelerate” the tech. It says they left after the company decided against launching the Meena LLM they’d built. Google acquired the Character.AI leadership team in August.
Advertisement
Character.AI’s website and mobile app has hundreds of custom AI chatbots, many modeled after popular characters from TV shows, movies, and video games. A few months ago, The Verge wrote about the millions of young people, including teens, who make up the bulk of its user base, interacting with bots that might pretend to be Harry Styles or a therapist. Another recent report from Wired highlighted issues with Character.AI’s custom chatbots impersonating real people without their consent, including one posing as a teen who was murdered in 2006.
Because of the way chatbots like Character.ai generate output that depends on what the user inputs, they fall into an uncanny valley of thorny questions about user-generated content and liability that, so far, lacks clear answers.
Character.AI has now announced several changes to the platform, with communications head Chelsea Harrison saying in an email to The Verge, “We are heartbroken by the tragic loss of one of our users and want to express our deepest condolences to the family.”
Some of the changes include:
Advertisement
“As a company, we take the safety of our users very seriously, and our Trust and Safety team has implemented numerous new safety measures over the past six months, including a pop-up directing users to the National Suicide Prevention Lifeline that is triggered by terms of self-harm or suicidal ideation,” Harrison said. Google didn’t immediately respond to The Verge’s request for comment.
An unseen, non-human hand moving the cursor across your computer screen and typing without using the keyboard in fiction is usually a sign of malicious AI hijacking something (or a friendly ghost helping you solve mysteries like the TV show Ghost Writer). Thanks to Anthropic’s new computer use feature for its AI assistant Claude, there’s a much more benevolent explanation now.
Fueled by an upgraded version of the Claude 3.5 Sonnet model, this AI – dubbed ‘computer use’ – lets you interact with your computer much like you would. It takes the AI assistant concept a step beyond text and a voice, with virtual hands typing, clicking, and otherwise manipulating your computer.
Anthropic bills computer use as a way for Claude to handle tedious tasks. It can help you fill out a form, search and organize information on your hard drive, and move information around. While OpenAI, Microsoft, and other developers have demonstrated similar ideas, Anthropic is the first to have a public feature, though it’s still in beta.
“With computer use, we’re trying something fundamentally new,” Anthropic explained in a blog post. Instead of making specific tools to help Claude complete individual tasks, we’re teaching it general computer skills—allowing it to use a wide range of standard tools and software programs designed for people.”
Advertisement
The computer use feature is due to Claude 3.5 Sonnet’s improved performance, particularly with digital tools and coding software. Though somewhat overshadowed by the spectacle of the computer use feature, Anthropic also debuted a new model called Claude 3.5 Haiku, a more advanced version of the lower-cost Anthropic model, though once capable of matching Anthropic’s previous highest performing model, Claude 3 Opus, while still being much cheaper.
Claude | Computer use for automating operations – YouTube
You can’t just give an order and walk away, either. Claude’s control of your computer has some technical troubles as well as deliberate constraints. On the technical side, Anthropic admitted Claude struggles with scrolling and zooming around a screen. That’s because the AI interprets what’s on your screen as a collection of screenshots, and then it tries to piece them together like a movie reel. Anything that happens too quickly or that changes perspective on the screen can flummox it. Still, Claude can do quite a lot by manipulating your computer, as seen above.
Unrestrained automation has obvious perils even when working perfectly, as so many sci-fi movies and books have explored. Claude isn’t Skynet, but Anthropic has placed restraints on the AI for more prosaic reasons. For instance, there are guardrails stopping Claude from interacting with social media or any government websites. Registering domain names or posting content is not allowed without human control.
Sign up for breaking news, reviews, opinion, top tech deals, and more.
“Because computer use may provide a new vector for more familiar threats such as spam, misinformation, or fraud, we’re taking a proactive approach to promote its safe deployment. We’ve developed new classifiers that can identify when computer use is being used and whether harm is occurring,” Anthropic wrote. “Learning from the initial deployments of this technology, which is still in its earliest stages, will help us better understand both the potential and the implications of increasingly capable AI systems.”
The Indian Cabinet recently approved four important space missions, marking a major step forward in India’s space exploration journey. These missions aim to boost India’s position in space research and technology, highlighting the country’s dedication to advancing its space capabilities and playing a key role in global space exploration efforts. The approved missions include Chandrayaan-4, the Venus Orbiter Mission, the creation of the Bharatiya Antariksha Station, and the Gaganyaan follow-up programme. Each mission comes with its own set of challenges and opportunities, and their success will depend on advanced technology, skilled expertise, and considerable financial support.
Foremost among them is the Chandrayaan-4 which is India’s next leap towards lunar exploration. Chandrayaan-4 is set to be a ground-breaking mission for India, focusing on collecting samples from the Moon and returning them to Earth. This mission, with an impressive budget of Rs 2,104.06 crore, builds upon the success of Chandrayaan-3, which proved India’s capability to land on the Moon. Now, Chandrayaan-4 aims to deepen our understanding of the Moon’s composition and geological history.
The primary objective of Chandrayaan-4 is to gather and analyse soil and rock samples from the lunar surface. These samples will provide valuable insights into the Moon’s structure and evolution, offering scientists a better understanding of its formation and the processes that have shaped it over billions of years.
“The mission is not without its challenges. A precise lunar landing is essential to ensure the spacecraft touches down safely in the targeted region. Additionally, efficient sample collection and the safe return of these samples to Earth are critical. These operations will demand high levels of precision and advanced technology to be successfully executed. To achieve its goals, Chandrayaan-4 will utilise advanced robotics and autonomous systems, which will allow the spacecraft to navigate the lunar surface and collect samples efficiently. A sophisticated heat shield will also be crucial to protect the spacecraft from the intense heat generated during re-entry into Earth’s atmosphere,” explained space expert Girish Linganna.
Advertisement
Chandrayaan-4 is more than just a sample return mission; it is a key stepping stone for India’s ambitious goal of landing astronauts on the Moon by 2040. Scheduled to launch in 2027, the mission will serve as a critical testing ground for technologies essential to crewed lunar missions. The mission will involve five modules launched in two separate phases, demonstrating ISRO’s innovative approach to handling complex space operations. These modules will work in tandem to collect lunar samples and return them to Earth—a feat previously achieved by only a few countries, including the United States and China.
“Chandrayaan-4 will test important capabilities, such as docking and undocking in lunar orbit, precision landings, and ensuring safe re-entry through Earth’s atmosphere. These technologies are vital for future crewed missions and will lay the groundwork for India’s eventual human moon landings,” remarked Linganna.
ISRO Chairman S. Somanath highlighted the mission’s importance, stating, “Chandrayaan-4 is not just about bringing back Moon rocks; it’s about demonstrating our capability to send humans to the Moon and safely return them. This mission represents a significant leap forward in India’s space ambitions, bringing the nation closer to achieving its goal of a human presence on the Moon.
Similarly the Venus Orbiter Mission (VOM) is set to launch in March 2028, with a budget of Rs 1,236 crore, recently approved by the Union Cabinet. This will be India’s second interplanetary mission, following the successful Mars Orbiter Mission, which was launched in 2013 and entered Mars’ orbit in 2014. The VOM aims to study Venus in detail, focusing on its surface features, clouds, lightning, volcanic activity, atmosphere, and how the Sun interacts with the planet. Scientists are particularly interested in understanding why Venus, once thought to be similar to Earth and possibly habitable, underwent such a drastic transformation.
Advertisement
This mission is expected to provide valuable answers to long-standing questions about Venus, shedding light on its evolution and contributing to important scientific discoveries. Out of the total budget of Rs 1,236 crore for the Venus Orbiter Mission, around Rs 824 crore will be allocated to the development of the spacecraft. ISRO will be in charge of designing and launching the spacecraft, ensuring the project is managed efficiently through their well-established procedures.
Various industries will be involved in building the spacecraft and launch vehicle, which is expected to create significant employment opportunities and have positive effects on other sectors through technological advancements.
The data collected from the mission will be shared with the scientific community using existing systems. Additionally, this mission will prepare India for future planetary missions that can carry larger scientific instruments.
“The Venus mission is highly intriguing but is expected to be one of the most difficult tasks ISRO has ever taken on. Venus is an extremely hostile planet, with its atmosphere mostly made up of carbon dioxide, resulting in scorching surface temperatures. Due to a strong greenhouse effect, temperatures can soar up to 870°F (470°C)—hot enough to melt lead. The planet’s surface is dotted with numerous volcanoes, many of which are dormant, though some could still be active. Adding to the challenge, Venus has an extremely high surface air pressure, about 90 times greater than what we experience at sea level on Earth,” pointed out Linganna.
Advertisement
Another major space mission is making significant strides in space exploration with its ambitious plan to build the Bharatiya Antariksha Station (BAS-1), the country’s first space station module. The Indian government has officially approved the construction of this module, which has been integrated into the expanded Gaganyaan mission. This crucial decision sets India on a path to operate its own space station by 2035 and also opens the door for a manned Moon mission by 2040.
The Bharatiya Antariksha Station (BAS) aims to create India’s very own space station, which will orbit 400 km above the Earth’s surface. Weighing 52 tonnes, this massive structure will act as a research hub where Indian astronauts and scientists can carry out experiments in areas like microgravity, astronomy, and Earth observation. The station will also allow astronauts to stay in orbit for periods of 15 to 20 days.
The Gaganyaan mission, originally designed for human spaceflight, has now been expanded to include eight missions, all to be completed by December 2028. This expansion covers not only additional uncrewed missions but also the launch of the BAS-1 unit, alongside the necessary hardware to support long-term human spaceflight activities.
With the updated scope, the overall budget for the Gaganyaan programme has risen to Rs 20,193 crore, with an additional Rs 11,170 crore allocated specifically for the expanded scope. This substantial financial investment underscores India’s growing ambition to become a major player in space exploration.
Advertisement
The Indian Space Research Organisation (ISRO) will spearhead the project, working in collaboration with industry partners, educational institutions, and various government agencies. The focus of the programme is to develop and demonstrate key technologies needed for extended human space missions. Four missions are planned under the current Gaganyaan schedule by 2026, followed by the construction of the BAS-1 module. Another four missions dedicated to technology demonstration and validation will be completed by 2028.
The Bharatiya Antariksha Station will play a pivotal role in advancing scientific research in microgravity, leading to breakthroughs that could benefit various sectors on Earth. The mission is also expected to create numerous job opportunities in high-tech fields related to space and technology.
Moreover, this initiative is poised to inspire a new generation of young Indians to pursue careers in science and technology. The innovations developed during this mission will have far-reaching societal benefits, further establishing India as a leader in space exploration and scientific discovery.
The fourth mission approved by the Union Cabinet focuses on the development of the Next Generation Launch Vehicle (NGLV). This new launch vehicle will be designed to carry heavier payloads, be cost-effective, and reusable, with the potential to be commercially successful. According to the government, the NGLV will have three times the payload capacity of the current LVM-3, while only costing 1.5 times more. Its reusability will help lower the cost of accessing space, and it will feature modular green propulsion systems. A total of Rs 8,240 crore has been approved for the NGLV, covering development costs, three test flights, the setup of necessary facilities, programme management, and the launch campaign.
Advertisement
According to the government, the NGLV will strengthen India’s ability to carry out a variety of space missions, such as human spaceflight, lunar exploration, and satellite launches. This will greatly enhance the country’s space programme and overall space ecosystem.
You must be logged in to post a comment Login