Microsoft announces new supercomputer, lays out vision for future AI work - Source
Gartner's 2025 Strategic Technology Trends Are Just Right
Every year Gartner releases its list of technology trends. This year the list looks like this (organized in three buckets):
Bucket #1: AI Imperatives & Risks
Bucket #2: New Frontiers of Computing
Bucket #3: Human Machine Synergy
What's good about this list is the buckets or "themes" that Gartner likes to call them. It's impossible not to group them together to explain larger trajectories of the technologies likely to impact all aspects of business. So organizing the trends as themes makes sense.
The AI theme is right – and my students would agree. In our classes on AI, Machine Learning and Generative AI and Emerging Business Technologies at the Villanova School of Business, we also track technology trends. We've moved from large language models to generative pre-trained transformers to custom GPTs to proprietary LLMs to Agentic AI pretty quickly noting how "easy" it is to create agents with tools like Autogen, LangChain and CrewAI, among others. There's no question that agentic AI is a major and significant trend in 2025 and beyond.
AI governance is necessary, but not always so easily defined or enforced. Gartner is correct to list this technology trend. But as it notes, "AI guidelines vary across regions and industries, making it difficult to establish consistent practices."
One of the major challenges of AI is its ability to create and disseminate disinformation. "Disinformation security" is not only an important trend, it's also a necessity.
The second theme – new frontiers of computing – maps nearly perfectly onto our Emerging Business Technologies course. Post-quantum cryptography – "data protection that is resistant to quantum computing (QC) decryption risks" – anticipates progress in quantum computing which we believe is inevitable and yet will at times be risky.
Our reaction to ambient invisible intelligence trends is "it's about time" we started talking about using technology to create natural, intuitive experiences.
Energy-efficient computing is a requirement that's exploding. While blockchain was the wake-up call about energy requirements, AI is the tsunami. Gartner is right to highlight this requirement regardless of how it's satisfied.
The hybrid computing trend is obvious.
The third theme – human machine synergy – is another one that's almost overdue. The first trend is spatial computing which resurrects augmented and virtual reality (AR/VR) to solve some of the problems we expected it to solve five years ago. The enabling technology is now ready to exploit AR/VR in areas that Gartner identifies, like "gaming, education and e-commerce … healthcare, retail and manufacturing."
While Gartner identifies the cost and unwieldy nature of head-mounted displays, we believe that these displays will shrink considerably in the next few years. The recent release of smart glasses from vendors like Meta and Viture Pro XR Glasses suggests how this will evolve.
Gartner identifies "polyfunctional robots" as an important trend and we agree. The marriage between robotics and AI has just begun and will soon create children capable of performing multiple tasks.
Over the years the Defense Advanced Research Projects Agency has invested in brain computer interfaces. In fact, decades ago, DARPA funded some of the earliest research in biocybernetics. Companies like Neuralink have already demonstrated significant potential here. Gartner correctly puts neurological enhancement on the list of technology trends.
Thematic OpportunitiesGartner's themes and trends map closely with the trends we see at the university. Our AI, Machine Learning and Generative AI and Emerging Business Technologies courses overlap with many of Gartner's trends.
We should follow Gartner's lead and organize them into baskets or themes. The figure below from our Emerging Business Technology course is organized around infrastructure (the inner circle) and applications technologies (the outer ring). The next course in AI, Machine Learning and Generative AI will focus heavily on agentic AI and the course on Emerging Business Technologies will focus heavily on sustainability, human machine synergy and robotics.
Business Technology Trends
SJ Andriole The Technology Trends BusinessGartner is not the only company that focuses on technology trends. Forrester, Deloitte, IBM, IDC, PWC and McKinsey all track and predict technology trends. The challenge of course is the volatility of these trends. The courses we teach at the university are, for example, redesigned every single semester. Five years ago generative AI was unrecognizable as an application platform. Today it's redefining industries. Where will it be five years from now? I'd like to see Gartner list the trends – and the industries most affected – 10 years from now. Or at least five years from now. It's much easier to do them year by year. Same for my courses. I'm not sure I could teach a course on Emerging Business Technologies that focuses on where the technologies will be in five years.
In addition to all this, trends forecasters should identify where the technologies will land across the vertical industries and the processes and sub-processes that define them. They should also probabilistically identify the processes and sub-processes that will be automated, improved or eliminated as AI, machine learning and generative AI evolve. This would enable strategists and tacticians to leverage technology in some competitive ways. One thing's for sure. No one can ignore technology trends this year, next year or any year.
This Louisiana Teacher Pushes His Peers To Embrace Tech: 'We Have To Teach For Tomorrow'
Suresh Chiruguru has long worked to bring computer science education to Louisiana classrooms.
Chiruguru, who recently was chosen as Louisiana's High School Teacher of the Year, teaches in Calcasieu Parish and serves on the board for the Louisiana Association of Computer Using Educators. He said his passion for computer science developed in 2012 after he received a grant that allowed him to purchase iPads, a TV and a laptop for his classroom.
Seeing how the devices made his students more engaged and gave them opportunities to tinker and explore, he wanted to do more to encourage their curiosity.
"At the time, there were no opportunities for students' independent learning," he said. By incorporating technology into his lessons, "they could come up with their own questions and search for the answers."
In 2022, Chiruguru received an International Society for Technology in Education certification, a highly regarded credential for educators who demonstrate a strong ability to incorporate technology into their lessons. The recognition inspired him to become an advocate for other teachers to embrace technology in their classrooms.
Since then, Chiruguru has helped lead Louisiana in its effort to modernize its computer science curriculum. He served on a task force that created the state's first computer-science learning standards for students in grades K-12, which the state board of education approved last month. The standards include topics that every student must learn, including about different computing systems and data analysis.
"We wanted to raise the bar for students to develop these digital literacy skills," he said, "so that they can be ready for the future."
We recently spoke with Chiruguru about why he thinks digital literacy is critical for today's students — and how schools can make sure students learn it.
This interview has been edited for length and clarity.
Why is it important to teach Louisiana students about computer science?
Computers are the future. We cannot just teach for today — we have to teach for tomorrow.
When we look at computer science education, that is one of the excellent pathways to lead students to a world-class education, and this is where we can break a lot of barriers and empower them with new skills.
Also, we are surrounded by so many chemical plants and factory industries in Louisiana. Our state needs highly digitally literate (job) candidates.
What are some of the biggest challenges schools face when preparing students to enter an increasingly digital workforce?
One of the [metrics] I use is the digital divide and how we're still struggling to close that.
The digital divide is the accessibility gap among different groups of students when it comes to having and being able to use technology. It could be due to socioeconomic factors, but whatever the reason, some groups of students may not be able to use certain technologies. We need to close this gap and provide an equitable education for every student.
In Louisiana, we now have our own computer science standards, but not every school is ready to embrace computer science education. It's going to take some time, and there needs to be support as far as finances are concerned. Not every school is fortunate enough to hire teachers to come in and implement these new initiatives.
What are some possible solutions to these challenges?
What the state has done is introduce the Computer Science Technology Assistance Grant, which helps districts purchase technology or software for students and pay for computer science teachers.
In Calcasieu, we have something called the Virtual Instruction Program. There are some courses that are not available at every school in the district, so qualified teachers are teaching from a central hub and students (in different schools) can access the classes (virtually). The students who are interested enroll and join the class with their computer from their school.
We're trying to push this program so that the courses are available for every student in the parish.
You're on the task force that helped develop the state's new computer science standards. Where do those stand now?
The standards were presented for public comment earlier this year and approved on Oct. 8. They can be viewed [on the state Department of Education website].
When I say computer science is a pathway to a world-class education, I mean an education where the digital divide is closed and every student, regardless of their background and socioeconomic status, is able to grow and develop valuable skills that open up future opportunities.
My philosophy with these standards is that they support our teachers and schools in preparing students for their career goals.
What's next for computer science in schools?
Schools will need to encourage students to be creative and innovative through technology, because that's where their ideas come to life.
They'll also need to encourage "growth mindsets." Learning to code, for example, requires a growth mindset because it involves a lot of trial and error — not only to find a solution for a problem, but to find the best solution.
When we have these wonderful things that come from a computer science education, any principal or any district can see what it has to offer students. I believe they're going to embrace it.
Anthropic's New AI Model Can Control Your PC
In a pitch to investors last spring, Anthropic said it intended to build AI to power virtual assistants that could perform research, answer emails, and handle other back-office jobs on their own. The company referred to this as a "next-gen algorithm for AI self-teaching" — one it believed that could, if all goes according to plan, automate large portions of the economy someday.
It took a while, but that AI is starting to arrive.
Anthropic on Tuesday released an upgraded version of its Claude 3.5 Sonnet model that can understand and interact with any desktop app. Via a new "Computer Use" API, now in open beta, the model can imitate keystrokes, button clicks, and mouse gestures, essentially emulating a person sitting at a PC.
"We trained Claude to see what's happening on a screen and then use the software tools available to carry out tasks," Anthropic wrote in a blog post shared with TechCrunch. "When a developer tasks Claude with using a piece of computer software and gives it the necessary access, Claude looks at screenshots of what's visible to the user, then counts how many pixels vertically or horizontally it needs to move a cursor in order to click in the correct place."
Developers can try out Computer Use via Anthropic's API, Amazon Bedrock, and Google Cloud's Vertex AI platform. The new 3.5 Sonnet without Computer Use is rolling out to Claude apps, and brings various performance improvements over the outgoing 3.5 Sonnet model.
Automating appsA tool that can automate tasks on a PC is hardly a novel idea. Countless companies offer such tools, from decades-old RPA vendors to newer upstarts like Relay, Induced AI, and Automat.
In the race to develop so-called "AI agents," the field has only become more crowded. AI agents remains an ill-defined term, but it generally refers to AI that can automate software.
Some analysts say AI agents could provide companies with an easier path to monetizing the billions of dollars that they're pouring into AI. Companies seem to agree: According to a recent Capgemini survey, 10% of organizations already use AI agents and 82% will integrate them within the next three years.
Salesforce made splashy announcements about its AI agent tech this summer, while Microsoft touted new tools for building AI agents yesterday. OpenAI, which is plotting its own brand of AI agents, sees the tech as a step toward super-intelligent AI.
Anthropic calls its take on the AI agent concept an "action-execution layer" that lets the new 3.5 Sonnet perform desktop-level commands. Thanks to its ability to browse the web (not a first for AI models, but a first for Anthropic), 3.5 Sonnet can use any website and any application.
Anthropic's new AI can control apps on a PC. Image Credits:Anthropic"Humans remain in control by providing specific prompts that direct Claude's actions, like 'use data from my computer and online to fill out this form'," an Anthropic spokesperson told TechCrunch. "People enable access and limit access as needed. Claude breaks down the user's prompts into computer commands (e.G. Moving the cursor, clicking, typing) to accomplish that specific task."
Software development platform Replit has used an early version of the new 3.5 Sonnet model to create an "autonomous verifier" that can evaluate apps while they're being built. Canva, meanwhile, says that it's exploring ways in which the new model might be able to support the designing and editing process.
But how is this any different than the other AI agents out there? It's a reasonable question. Consumer gadget startup Rabbit is building a web agent that can do things like buying movie tickets online; Adept, which was recently acqui-hired by Amazon, trains models to browse websites and navigate software; and Twin Labs is using off-the-shelf models, including OpenAI's GPT-4o, to automate desktop processes.
Anthropic claims the new 3.5 Sonnet is simply a stronger, more robust model that can do better on coding tasks than even OpenAI's flagship o1, per the SWE-bench Verified benchmark. Despite not being explicitly trained to do so, the upgraded 3.5 Sonnet self-corrects and retries tasks when it encounters obstacles, and can work toward objectives that require dozens or hundreds of steps.
The new Claude 3.5 Sonnet model's performance on various benchmarks. Image Credits:AnthropicBut don't fire your secretary just yet.
In an evaluation designed to test an AI agent's ability to help with airline booking tasks, like modifying a flight reservation, the new 3.5 Sonnet managed to complete less than half of the tasks successfully. In a separate test involving tasks like initiating a return, 3.5 Sonnet failed roughly a third of the time.
Anthropic admits the upgraded 3.5 Sonnet struggles with basic actions like scrolling and zooming, and that it can miss "short-lived" actions and notifications because of the way it takes screenshots and pieces them together.
"Claude's Computer Use remains slow and often error-prone," Anthropic writes in its post. "We encourage developers to begin exploration with low-risk tasks."
Risky businessBut is the new 3.5 Sonnet capable enough to be dangerous? Possibly.
A recent study found that models without the ability to use desktop apps, like OpenAI's GPT-4o, were willing to engage in harmful "multi-step agent behavior," such as ordering a fake passport from someone on the dark web, when "attacked" using jailbreaking techniques. Jailbreaks led to high rates of success in performing harmful tasks even for models protected by filters and safeguards, according to the researchers.
One can imagine how a model with desktop access could wreak more havoc — say, by exploiting app vulnerabilities to compromise personal info (or storing chats in plaintext). Aside from the software levers at its disposal, the model's online and app connections could open avenues for malicious jailbreakers.
Anthropic doesn't deny that there's risk in releasing the new 3.5 Sonnet. But the company argues that the benefits of observing how the model is used in the wild ultimately outweigh this risk.
"We think it's far better to give access to computers to today's more limited, relatively safer models," the company wrote. "This means we can begin to observe and learn from any potential issues that arise at this lower level, building up computer use and safety mitigations gradually and simultaneously."
Image Credits:AnthropicAnthropic also says it has taken steps to deter misuse, like not training the new 3.5 Sonnet on users' screenshots and prompts, and preventing the model from accessing the web during training. The company says it developed classifiers to "nudge" 3.5 Sonnet away from actions perceived as high-risk, such as posting on social media, creating accounts, and interacting with government websites.
As the U.S. General election nears, Anthropic says it is focused on mitigating election-related abuse of its models. The U.S. AI Safety Institute and U.K. Safety Institute, two separate but allied government agencies dedicated to evaluating AI model risk, tested the new 3.5 Sonnet prior to its deployment.
Anthropic told TechCrunch it has the ability to restrict access to additional websites and features "if necessary," to protect against spam, fraud, and misinformation, for example. As a safety precaution, the company retains any screenshots captured by Computer Use for at least 30 days — a retention period that might alarm some devs.
We asked Anthropic under which circumstances, if any, it would hand over screenshots to a third party (e.G. Law enforcement) if asked. A spokesperson said that the company would "comply with requests for data in response to valid legal process."
"There are no foolproof methods, and we will continuously evaluate and iterate on our safety measures to balance Claude's capabilities with responsible use," Anthropic said. "Those using the computer-use version of Claude should take the relevant precautions to minimize these kinds of risks, including isolating Claude from particularly sensitive data on their computer."
Hopefully, that'll be enough to prevent the worst from occurring.
A cheaper modelToday's headliner might've been the upgraded 3.5 Sonnet model, but Anthropic also said an updated version of Haiku, the cheapest, most efficient model in its Claude series, is on the way.
Claude 3.5 Haiku, due in the coming weeks, will match the performance of Claude 3 Opus, once Anthropic's state-of-the-art model, on certain benchmarks at the same cost and "approximate speed" of Claude 3 Haiku.
"With low latency, improved instruction following, and more accurate tool use, Claude 3.5 Haiku is well suited for user-facing products, specialized sub-agent tasks, and generating personalized experiences from huge volumes of data–like purchase history, pricing, or inventory data," Anthropic wrote in a blog post.
3.5 Haiku will initially be available as a text-only model and later as part of a multimodal package that can analyze both text and images.
3.5 Haiku's benchmark performance. Image Credits:AnthropicSo once 3.5 Haiku is available, will there be much reason to use 3 Opus? What about 3.5 Opus, 3 Opus' successor, which Anthropic teased back in June?
"All of the models in the Claude 3 model family have their individual uses for customers," the Anthropic spokesperson said. "Claude 3.5 Opus is on our roadmap and we'll be sure to share more as soon as we can."
TechCrunch has an AI-focused newsletter! Sign up here to get it in your inbox every Wednesday.
Comments
Post a Comment