Over the course of the year, we’ve had the pleasure of working with many talented researchers from across the globe. As 2025 draws to a close, we take a look back at some of the excellent blog posts, interviews and podcasts from our contributors.
Teaching robot policies without new demonstrations: interview with Jiahui Zhang and Jesse Zhang
Jiahui Zhang and Jesse Zhang to tell us about their framework for learning robot manipulation tasks solely from language instructions without per-task demonstrations.
CoRL2025 – RobustDexGrasp: dexterous robot hand grasping of nearly any object
Hui Zhang writes about work presented at CoRL2025 on RobustDexGrasp, a novel framework that tackles different grasping challenges with targeted solutions.
Robot Talk Episode 133 – Creating sociable robot collaborators, with Heather Knight
Robot Talk host Claire Asher chatted to Heather Knight from Oregon State University about applying methods from the performing arts to robotics.
Generations in Dialogue: Human-robot interactions and social robotics with Professor Marynel Vasquez
In this podcast from AAAI, host Ella Lan asked Professor Marynel Vázquez about what inspired her research direction, how her perspective on human-robot interactions has changed over time, robots navigating the social world, and more.
Learning robust controllers that work across many partially observable environments
In this blog post, Maris Galesloot summarizes work presented at IJCAI 2025, which explores designing controllers that perform reliably even when the environment may not be precisely known.
Robot Talk Episode 130 – Robots learning from humans, with Chad Jenkins
Claire Asher chatted to Chad Jenkins from University of Michigan about how robots can learn from people and assist us in our daily lives.
Interview with Zahra Ghorrati: developing frameworks for human activity recognition using wearable sensors
Zahra Ghorrati is pursuing her PhD at Purdue University, where her dissertation focuses on developing scalable and adaptive deep learning frameworks for human activity recognition (HAR) using wearable sensors.
Self-supervised learning for soccer ball detection and beyond: interview with winners of the RoboCup 2025 best paper award
We caught up with some of the authors of the RoboCup 2025 best paper award to find out more about the work, how their method can be transferred to applications beyond RoboCup, and their future plans for the competition.
#IJCAI2025 distinguished paper: Combining MORL with restraining bolts to learn normative behaviour
Agata Ciabattoni and Emery Neufeld introduce a framework for guiding reinforcement learning agents to comply with social, legal, and ethical norms.
Robot Talk Episode 114 – Reducing waste with robotics, with Josie Gotz
Claire Asher chatted to Josie Gotz from the Manufacturing Technology Centre about robotics for material recovery, reuse and recycling.
Multi-agent path finding in continuous environments
Kristýna Janovská and Pavel Surynek write about how can a group of agents minimise their journey length whilst avoiding collisions.
RoboCupRescue: an interview with Adam Jacoff
Find out what’s new in the RoboCupRescue League this year.
An interview with Nicolai Ommer: the RoboCup Soccer Small Size League
We caught up with Nicolai to find out more about the Small Size League, how the auto referees work, and how teams use AI.
Interview with Kate Candon: Leveraging explicit and implicit feedback in human-robot interactions
Hear from PhD student Kate about her work on human-robot interactions.
AIhub coffee corner: Agentic AI
The AIhub coffee corner captures the musings of AI experts over a short conversation.
Generations in Dialogue: Multi-agent systems and human-AI interaction with Professor Manuela Veloso
Host Ella Lan chats to Professor Manuela Veloso about her research journey and path into AI, the history and evolution of AI research, inter-generational collaborations, and more.
Preparing for kick-off at RoboCup2025: an interview with General Chair Marco Simões
We spoke to Marco Simões, one of the General Chairs of RoboCup 2025 and President of RoboCup Brazil.
Gearing up for RoboCupJunior: Interview with Ana Patrícia Magalhães
RoboCup Junior Rescue @ WK RoboCup 2024. Photo: RoboCup/Bart van Overbeeke
Shirking its ‘fun toy’ image, AI like ChatGPT was increasingly seen by the business community in Q4 as a must-have productivity tool destined to reward early adopters and punish Luddites.
ChatGPT’s maker, for example, released a study finding that everyday AI business users are saving at least 40 minutes a day on busy work – while power users are saving up to two hours-a-day.
Meanwhile, MIT released a report concluding that AI can currently eliminate 12% of all jobs — as more businesses were found issuing decrees along the lines of ‘Use AI or You’re Fired.”
Especially alarming for writers was a move by media outlet Business Insider, which started publishing news stories completely written by AI and carrying an AI byline.
Plus, photographers and graphic artists got their own dose of rubber-meets-road reality with new, back-to-back releases of extremely powerful new AI imaging tools built into Gemini and ChatGPT.
But despite the breakneck development, users also continued to report that AI agents – designed to automate multi-step tasks – are continuing to fail miserably.
Plus, many users continued to ‘forget’ that AI makes-up facts, and that using AI responses without fact-checks can lead to major ‘egg-on-face’ moments.
The most gleaming ray of hope in al lthis: The Wall Street Journal reported that AI is considered so essential to U.S. defense, there’s a good chance the U.S. government will bail-out the AI industry if the much-feared ‘AI Bubble’ bursts.
Here’s a full rundown of how those stories — and more — helped shape AI writing in Q4 2024:
*ChatGPT-Maker Study: The State of Enterprise AI: New research from OpenAI finds that everyday business users of AI are saving about 40-60 minutes-a-day when compared to working without the tool.
Even better, the heaviest AI users say they’re saving up to two hours a day with the tech.
*AI Can Already Eliminate 12% of U.S. Workforce: A new study from MIT finds that AI can already eliminate 12% of everyday jobs.
Dubbed the “Iceberg Index,” the study simulated AI’s ability to handle – or partially handle – nearly 1,000 occupations that are currently worked by more than 150 million in the U.S.
Observes writer Megan Cerullo: “AI is also already doing some of the entry-level jobs that have historically been reserved for recent college graduates or relatively inexperienced workers.”
*Use AI or You’re Fired: In another sign that the days of ‘AI is Your Buddy’ are fading fast, increasing numbers of businesses have turned to strong-arming employees when it comes to AI.
Observes Wall Street Journal writer Lindsay Ellis: “Rank-and-file employees across corporate America have grown worried over the past few years about being replaced by AI.
“Something else is happening now: AI is costing workers their jobs if their bosses believe they aren’t embracing the technology fast enough.”
*Breaking News Gets an AI Byline at Business Insider: The next news story you read from Business Insider may be completely written by AI — and carry an AI byline.
The media outlet has announced a pilot test of a story writing algorithm that will grab a piece of breaking news and give it context by combining it with data drawn from stories in the Business Insider archive.
The only human involvement will be an editor, who will look over the finished product before it’s published.
*Study: AI Agents Virtually Useless at Completing Freelance Assignments: New research finds that much-ballyhooed AI agents are literally horrible at completing everyday assignments found on freelance brokerage sites like Fiverr and Upwork.
Observes writer Frank Landymore: “The top performer, they found, was an AI agent from the Chinese startup Manus with an automation rate of just 2.5 percent — meaning it was only able to complete 2.5 percent of the projects it was assigned at a level that would be acceptable as commissioned work in a real-world freelancing job, the researchers said.
“Second place was a tie, at 2.1 percent, between Elon Musk’s Grok 4 and Anthropic’s Claude Sonnet 4.5.”
*Oops, Sorry Australia, Here’s Your Money Back: Consulting firm Deloitte has agreed to refund the Australian government $440,000 for a study both agree was riddled by errors created by AI.
Observes writer Krishani Dhanji: “University of Sydney academic Dr. Christopher Rudge — who first highlighted the errors — said the report contained ‘hallucinations’ where AI models may fill in gaps, misinterpret data, or try to guess answers.”
Insult to injury: The near half-million-dollar payment is only a partial refund to what the Australian government actually paid for the flawed research.
*Forget Benchmarks: Put AI Through Your Own Tests Before You Commit: While benchmarks offer an indication of the AI solution you’re considering, you really need to put the AI through your own tests before you opt for anything, according to Ethan Mollick.
An associate professor of management at the Wharton School of the University of Pennsylvania, Mollick studies and teaches entrepreneurship, innovation and how AI is changing work and education.
Observes Mollick: “You need to know specifically what your AI is good at — not what AIs are good at on average.”
*AI Gets a Number One Country Hit: Well, it’s official: AI can now write and produce a country hit with the best of ’em.
“Walk My Walk,” a song credited to an AI artist named ‘Breaking Rust,’ has hugged the number one spot on Billboard’s Country Digital Song Sales chart for two weeks in a row.
The hit comes on the heels of another AI hit in another music genre, according to Billboard’s Adult R&B Airplay chart.
*Free AI from China Keeps U.S. Tech Titans on Their Toes: While still holding a slim lead, major AI players like ChatGPT, Gemini and Claude are feeling the nip-at-their-heels of ‘nearly as good’ – and free – AI alternatives from China.
Key Chinese players like DeepSeek and Qwen, for example, are within chomping distance of the U.S. marketing leaders — and are Open Source, or freely available for download and tinkering.
One caveat: Researchers have found AI code embedded in some Chinese AI that can be used to forward your data along to the Chinese Communist Party.
*Solution to AI Bubble Fears: U.S. Government?: The Wall Street Journal reports that AI is now considered so essential to U.S. defense, the U.S. government may step in to save the AI industry — should it implode from the irrational exuberance of investors.
Observes lead writer Sarah Myers West: “The federal government is already bailing out the AI industry with regulatory changes and public funds that will protect companies in the event of a private sector pullback.
“Despite the lukewarm market signals, the U.S. government seems intent on backstopping American AI — no matter what.”
Share a Link: Please consider sharing a link to https://RobotWritersAI.com from your blog, social media post, publication or emails. More links leading to RobotWritersAI.com helps everyone interested in AI-generated writing.
–Joe Dysart is editor of RobotWritersAI.com and a tech journalist with 20+ years experience. His work has appeared in 150+ publications, including The New York Times and the Financial Times of London.
The post Top Ten Stories in AI Writing, Q4 2025 appeared first on Robot Writers AI.
Reports surfaced this week confirming what many industry observers had long suspected but feared to articulate: China has launched a “Manhattan Project” for semiconductors. This massive, state-backed initiative has a singular, existential goal—to reverse-engineer the ultra-complex lithography machines currently monopolized […]
The post The Silicon Manhattan Project: China’s Atomic Gamble for AI Supremacy appeared first on TechSpective.
Most enterprises scaling agentic AI are overspending without knowing where the capital is going. This isn’t just a budget oversight. It points to deeper gaps in operational strategy. While building a single agent is a common starting point, the true enterprise challenge is managing quality, scaling use cases, and capturing measurable value across a fleet of 100+ agents.
Organizations treating AI as a collection of isolated experiments are hitting a “production wall.” In contrast, early movers are pulling ahead by building, operating, and governing a mission-critical digital agent workforce.
New IDC research reveals the stakes:
The competitive gap is no longer about build speed. It is about who can operate a safe, “Tier 0” service foundation in any environment.
The “hidden AI tax” is not a one-time fee; it is a compounding financial drain that multiplies as you move from pilot to production. When you scale from 10 agents to 100, a lack of visibility and governance turns minor inefficiencies into an enterprise-wide cost crisis.
The true cost of AI is in the complexity of operation, not just the initial build. Costs compound at scale due to three specific operational gaps:
Moving from a pilot to production is a structural leap. Challenges that seem manageable in a small experiment compound exponentially at scale, leading to a production wall where technical debt and operational friction stall progress.
Teams face a hidden burden maintaining zero downtime in mission-critical environments. In high-stakes industries like manufacturing or healthcare, a single failure can stop production lines or cause a network outage.
Example: A manufacturing firm deploys an agent to autonomously adjust supply chain routing in response to real-time disruptions. A brief agent failure during peak operations causes incorrect routing decisions, forcing multiple production lines offline while teams manually intervene.
Cloud vendors typically lock organizations into specific environments, preventing deployment on-premises, at the edge, or in air-gapped sites. Enterprises need the ability to maintain AI ownership and comply with sovereign AI requirements that cloud vendors cannot always meet.
Example: A healthcare provider builds a diagnostic agent in a public cloud, only to find that new Sovereign AI compliance requirements demand data stay on-premises. Because their architecture is locked, they are forced to restart the entire project.
Teams are overwhelmed by “infrastructure plumbing” and struggle to validate or scale agents as models and tools constantly evolve. This unsustainable burden distracts from developing core business requirements that drive value.
Example: A retail giant attempts to scale customer service agents. Their engineering team spends weeks manually stitching together OAuth, identity controls, and model APIs, only to have the system fail when a tool update breaks the integration layer.
Connecting inference serving with runtimes is complex, often driving up compute costs and failing to meet strict latency requirements. Without efficient runtime orchestration, organizations struggle to balance performance and value in real time.
Example: A telecommunications firm deploys reasoning agents to optimize network traffic. Without efficient runtime orchestration, the agents suffer from high latency, causing service delays and driving up costs.
For 68% of organizations, clarifying risk and compliance implications is the top requirement for agent use. Without this clarity, governance becomes the single biggest obstacle to expanding AI.
Success is no longer defined by how fast you experiment, but by your ability to focus on productionizing an agentic workforce from the start. This requires AI-first governance that enforces policy, cost, and risk controls at the agent runtime level, rather than retrofitting guardrails after systems are already live.
Example: A company uses an agent for logistics. Without AI-first governance, the agent might trigger an expensive rush-shipping order through an external API after misinterpreting customer frustration. This results in a financial loss because the agent operated without a policy-based safeguard or a “human-in-the-loop” limit.
This productionization-focused approach to governance highlights a key difference between platforms designed for agentic systems and those whose governance remains limited to the underlying data layer.
The 100-agent mark is where the gap between early movers and the rest of the market becomes a permanent competitive divide. Closing this gap requires a unified platform approach, not a fragmented stack of point tools.
Platforms built for managing an agentic workforce are designed to address the operational challenges that stall enterprise AI at scale. DataRobot’s Agent Workforce Platform reflects this approach by focusing on several foundational capabilities:
The competitive gap is widening. Early movers who invest in a foundation of governance, unified tooling, and cost visibility from day one are already pulling ahead. By focusing on the digital agent workforce as a system rather than a collection of experiments, you can finally move beyond the pilot and deliver real business impact at scale.
Want to learn more? Download the research to discover why most AI pilots fail and how early movers are driving real ROI. Read the full IDC InfoBrief here.
The post The 100-agent benchmark: why enterprise AI scale stalls and how to fix it appeared first on DataRobot.