From the emergence of autonomous AI agents capable of executing complex tasks to the integration of cutting-edge AI capabilities directly into devices, New Year 2025 promises to accelerate AI innovation at an unprecedented pace. The rise of humanoid robots, equipped with advanced AI systems, hints at a future where robots seamlessly coexist with humans, tackling labour-intensive tasks. Meanwhile, cost-efficient advancements in AI technologies aim to make these breakthroughs more accessible, addressing the challenges of high operational expenses.
Also read: OpenAI in 2024: ChatGPT Search, Sora AI video, and all the big wins
Industry giants like OpenAI, Google, Microsoft, and NVIDIA are leading the charge, unveiling transformative technologies that will profoundly influence how we live and work. As these trends converge, 2025 is set to be a landmark year, ushering in a wave of AI-driven possibilities that will redefine our relationship with technology. Let’s take a closer look…
If 2022 was the birth of AI chatbots as we know it, thanks to OpenAI’s ChatGPT, then by all indications 2025 will see a lot of AI agents coming out into the open from their current secretive research bubble – and no I’m not talking about the garden variety agents referenced in The Matrix! How they will change our world is anyone’s guess at this point, but the dawn of AI agents certainly promises to inject some excitement into the AI landscape that’s becoming – dare I say – more drab by the day.
In the last two years, the world has seen a lot of breakneck advancement in the Generative AI space, right from text-to-text, text-to-image and text-to-video based Generative AI capabilities. And all of that’s been nothing short of stepping stones for the next big AI breakthrough – AI agents.
Also read: Elon Musk in 2024: Top Tesla, SpaceX, AI and Neuralink highlights
In an interview with MIT Technology Review, OpenAI’s Sam Altman described the best application of AI as a “super-competent colleague that knows absolutely everything about my whole life, every email, every conversation I’ve ever had, but doesn’t feel like an extension.” That’s what an AI agent will look like, in essence. Are you ready for J.A.R.V.I.S., Tony Stark’s intelligent AI assistant from Iron Man, or Samantha from Her, a significantly advanced AI operating system than what we’ve experienced till now?
Simply put, an AI agent is a slightly more advanced AI program that can perform certain autonomous tasks that aren’t just limited to its own base program. ChatGPT or Gemini can write code for you, if you ask for it, but it can’t go and create a website or an app from that code, where the website is live with a domain name or the app published on the app store. An AI agent will be able to do these things – I’m not saying these exact tasks that I suggested above, but these AI agents will have the ability to not just show what needs to be done but also go ahead and do some of that work.
According to Amazon’s official AWS blog, humans will set broad goals for any given AI-related task, where the AI agent will independently choose the best actions it needs to perform to achieve those goals. Amazon further explains how in a customer service scenario, a future AI agent will automatically try to satisfy a calling customer’s query – by looking up internal information, by asking different questions to the human customer, by taking stock of the situation and responding with a solution that solves the calling customer’s problem. In this scenario, the AI agent handles the customer’s call on its own – without passing the call to a human customer support expert. In fact, whether or not to transfer a call to a human customer support expert is determined automatically by the AI agent.
AI agents will be superior to simple AI chatbots thanks to their advanced reasoning capabilities, suggests IBM’s blog post on AI agents. Unlike traditional AI chatbots like ChatGPT or Gemini, which give highly scripted responses to user queries, AI agents will have the ability to plan, think through and adapt to new information, enabling them to handle much more complex tasks with minimal human intervention or supervision.
All the big movers and shakers of the AI industry are planning to release their version of AI agents for the public very soon in 2025, if they haven’t done it already by late 2024. OpenAI’s reportedly working on getting their AI agent, codenamed Operator, out into the open for everyone to check out by January 2025. The AI agent is expected to be capable of autonomously operating certain tasks within your computer, like booking flight tickets and implementing code, among other things. Google seems to be working on several AI agent projects, one of which is known as Project Jarvis, which recently leaked on the Chrome Web Store. This AI agent will supposedly reside within Google Chrome browser, with the ability to not only automate and execute tasks within the browser but also operate other apps on the host PC or computer. While there’s no set release date for Project Jarvis yet, however, Google’s Gemini 2.0 AI model is expected to have AI agents built-in to offer enhanced capabilities – it’s expected to release later this year in 2024.
Microsoft is also working aggressively on AI agents, something that it had announced earlier in the year in 2024. According to its official blog, new capabilities in Copilot Studio will allow Microsoft customers to create powerful autonomous AI agents. Some of these demonstrations are in public preview at the moment, where AI agents can draw upon work or business data from different Microsoft Office 365 apps to undertake a variety of assistive tasks – like IT help desk, employee onboarding, coordinating sales and service, and more.
Anthropic, an AI startup competing with OpenAI, has already released its AI agent for people to try. According to Techcrunch, Anthropic has made significant upgrades to its Claude 3.5 Sonnet AI model which now lets it use the host computer – yes, it can interact with computers in a way that mimics humans. It can move the cursor around the screen, click on apps and buttons, and it can potentially interact with other softwares and programs installed in your PC to autonomously execute various tasks. How scary and cool is that?!
If the world wasn’t prepared for Generative AI back in 2022, then let me tell you it’s certainly not prepared for AI agents and all the various ways it can impact our lives – for better or worse.
On-device AI refers to artificial intelligence models that run directly on your device (like your smartphone, tablet, or laptop) rather than relying on remote servers. This means tasks are processed locally, offering several advantages: speed of AI execution which can operate completely offline and thereby enhance data privacy. Everything from smarter assistants on smartphones, real-time language translation in speech, and more capable AI applications with enhanced local device context is going to appear on everything from smartphones to laptops and then some, according to Intel, Qualcomm, Samsung and other chip and device makers.
Qualcomm has been a strong advocate for on-device AI, emphasising its role in delivering efficient and responsive user experiences. At the Snapdragon Summit 2024, the company showcased AI-driven innovations for mobile and automotive platforms, highlighting the importance of on-device processing in reducing latency and improving performance.
Also read: AI chips in 2024: Top 5 AI laptops, smartphone and AI hardware milestones
An interesting trend report by Gartner suggests that by 2026, as much as 30% of workers will utilise digital “charisma filters” to enhance their professional interactions and career progression. These filters, powered by on-device generative AI, suggest a trend toward embedding AI functionalities within personal devices to improve user experiences. Let me tell you, Gartner isn’t alone in this assessment.
Deloitte predicts that by the end of 2025, over 30% of smartphones shipped will be equipped with location generative AI capabilities. This integration is expected to transform user interactions and drive substantial growth in the mobile device market. In a similar vein, IDC forecasts a 73.1% increase in generative AI capable smartphone shipments in 2025, followed by consistent double-digit growth in subsequent years. On the PC front, according to Canalys, AI-capable PC shipments are projected to surpass 100 million units in 2025, accounting for 40% of all PC shipments. This indicates a substantial shift towards integrating AI functionalities directly into personal computing and mobile devices going forward.
The year 2025 is poised to be a pivotal milestone in the evolution of AI-powered humanoid robot systems, marking significant advancements in their development and integration across various sectors. According to a report by Goldman Sachs, the global market for humanoid robots is projected to reach $38 billion by 2035, with substantial growth anticipated to commence around 2025. This surge is attributed to rapid advancements in artificial intelligence, particularly in large language models and end-to-end AI systems, which enhance the capabilities and versatility of humanoid robots.
A company called Figure AI is focused on creating general-purpose humanoid robots capable of performing a variety of tasks. Their latest model, Figure 02, unveiled in August 2024, features a sleek design with integrated cabling, enhanced battery capacity, and advanced AI capabilities for conversational interactions. The robot is currently being tested in BMW’s manufacturing facilities.
Similarly, Clone Robotics, a Polish company, has developed a humanoid robot with water-powered muscles that mimic human movements and strength. 1X is a Norwegian company developing humanoid robots to assist humans in everyday tasks. Their robot, EVE, is designed for roles in security, healthcare, and other service-oriented industries, focusing on safety and affordability. Chinese electric vehicle manufacturer Xpeng has introduced the “Iron” robot, a six-foot-tall humanoid designed with advanced AI and mechanical capabilities. This development is part of Xpeng’s broader strategy to integrate robotics into their product lineup, including innovations like a new hybrid system and a modular flying car.
Also read: AI humanoids in 2024: Top robotics updates from Tesla to NVIDIA
Elon Musk, CEO of Tesla, has been vocal about the company’s ambitions in humanoid robotics. In July 2024, Musk announced that Tesla’s humanoid robot, Optimus, would enter low-volume production for internal use by early 2025, with plans for broader deployment later in the year through 2026. He emphasised that these robots are designed to perform tasks that are “boring, repetitive, and dangerous,” aiming to enhance productivity and safety in various environments. Musk envisions a future where humanoid robots are commonplace, assisting in both industrial and domestic settings. He has highlighted the potential for these robots to revolutionise labour markets and address workforce shortages by taking on tasks that are undesirable or hazardous for humans.
Similarly, Jensen Huang, CEO of NVIDIA, has expressed a strong commitment to advancing AI and robotics technologies. At NVIDIA’s GTC 2024 conference, Huang discussed the company’s efforts in developing AI platforms that support humanoid robots. He introduced the Jetson Thor computer, designed to enhance the capabilities of humanoid robots by providing advanced AI processing power. This technology aims to improve robots’ autonomy and their ability to interact seamlessly with human environments.
Huang has emphasised the importance of creating AI models that enable robots to understand and navigate complex human settings. He highlighted NVIDIA’s work on simulation platforms that allow for the training and testing of robots in virtual environments before deployment in the real world. This approach is intended to accelerate the development of safe and efficient humanoid robots capable of performing a wide range of tasks alongside humans.
As these technologies evolve, the collaboration between hardware development, as seen with Tesla’s Optimus, and AI processing capabilities, as provided by NVIDIA, will be crucial. The integration of sophisticated AI with advanced robotics hardware is expected to lead to more capable and versatile humanoid robots, transforming industries and daily life by 2025 and beyond.
While we all absolutely love using GenAI tools like ChatGPT, Gemini, Copilot and others, let me tell you that training and operating large-scale generative AI models, such as those developed by OpenAI, Google, and Anthropic, are an expensive business and not everyone’s cup of tea. For instance, OpenAI’s GPT-4 model, released in 2023, incurred an estimated training cost of $78.4 million. Beyond initial training, the operational costs of deploying these models are considerable as well. For example, inference – the process of running trained models to generate outputs – requires substantial computational resources, leading to constant ongoing expenses. For example, Dario Amodei, CEO of Anthropic, noted that training AI models can cost around $100 million at the lower end, with some models in development reaching up to a billion dollars.
Also read: How LLMs revolutionised our AI journey in 2024
This doesn’t take into account the cost of AI hardware, which can truly make your eyes water. Just take a look at NVIDIA’s Blackwell components’ costs. The GB200 Grace Blackwell superchips can be as much $70,000 each, while the Blackwell B100 accelerator isn’t exactly cheap at $30,000 to $35,000. Server racks with multiple of these AI workload chips running together can easily go over a million dollars, if not more. The Blackwell chips are expected to be available for volume sales in early 2025. They’re designed to be a scalable architecture that can be expanded into a supercomputer. The chips are expected to deliver breakthroughs in performance and efficiency, addressing the growing demands in AI, data centres, and gaming technologies. But as you can see, they are atrociously expensive. These figures highlight the financial challenges faced by organisations in sustaining and scaling generative AI technologies.
The cost of training these large AI models has been a significant concern for AI companies – however, several developments suggest that these costs are expected to decrease by 2025. Companies like Amazon Web Services (AWS) are investing in custom AI chips, such as Trainium, to offer more cost-effective alternatives to traditional GPUs. AWS announced plans to provide $110 million in free computing credits to AI researchers, aiming to challenge NVIDIA’s dominance and reduce training expenses.
The proliferation of open-source AI models like Meta’s Llama allows organisations to leverage pre-trained models, significantly cutting down on the resources required for training from scratch. This approach democratises access to advanced AI capabilities and reduces associated costs. Innovations in training methodologies, such as model distillation and transfer learning, enable the development of smaller, more efficient models that require less computational power and time to train, thereby lowering costs.
Also read: How SLMs made AI more compact and powerful in 2024
