What does it take for a robot to hold a conversation with a room, not just a person? That was one of the questions at the heart of my last week (20th-24th July) at the Imperial Robotics Summer School, hosted at Imperial College, London; a week that sharpened my thinking and pushed me to look at robotics problems from angles I don't usually get to in my day-to-day work. The summer school brought together emerging researchers, academics, and industry professionals from across the robotics community for an intensive, hands-on programme. I came away having learned a huge amount, from expert-led lectures on robot kinematics, dynamics, sensing and control, and robot learning, through to specialised sessions on aerial robotics, robot intelligence, surgical robot vision, bio-inspired sensing and control, and personal assistive robotics. One of the standout parts of the week was getting into Imperial's robotics labs themselves, seeing research up close across adaptive and intelligent robotics, aerial robotics, robotic surgery, manipulation and touch, and assistive robotics.

Rhys, right, was sent this photo of himself taken at a Melbourne cafe by a stranger wearing Meta's Ray-Ban glasses Photograph: Supplied Rhys, right, was sent this photo of himself taken at a Melbourne cafe by a stranger wearing Meta's Ray-Ban glasses Photograph: Supplied This man was secretly snapped by someone with smartglasses. The backlash to what some dub'pervert glasses' is growing - but people living with disabilities point out the new tech can also be life-changing Fri 7 Aug 2026 11.00 EDTLast modified on Sat 8 Aug 2026 01.47 EDT One summer day earlier this year, Rhys left a cafe in Melbourne where he had been working remotely, when his phone pinged with a message on the dating app Grindr. The message was a photo of him sitting at a table outside the cafe on his laptop, with his dog seated on the ground next to him. In the foreground, in a man's hand was an iPhone unlocked. The photo had been taken moments before with Meta's Ray-Ban glasses, unbeknown to Rhys.

AI is like ultra-processed food for developing brains. T he human brain has evolved to learn from specific instructors. When a child hears a gentle spoken word or a soothing lullaby from another person, when their cry is responded to, their brain lights up. Nurturing interactions, or what child-development experts call "serve and return" exchanges, help shape and strengthen the 1 million new neural connections taking place in a child's brain every single second. Every time a child cries or serves up a bid for attention and receives a caregiver's response, that back-and-forth helps build the circuits underlying their communication, learning, and emotional well-being.

Imagine you want to teach a robot to push an object on a table. The standard recipe in robot learning is to collect hundreds of expert demonstrations on a real robot, train an imitation learning policy on that data, and then evaluate the policy by running it many times on the same real robot. Both stages (data collection and evaluation) are slow, expensive, and hard to reproduce: hardware breaks, lighting changes, objects drift out of place, and every new task means more hours in the lab. A natural question is whether we can replace some of this real-robot work with a simulator. Classical physics-based simulators are powerful, but building one for a new task means manually modeling geometries, contacts, friction, and deformation, and the resulting simulator often still does not match reality closely enough for policies trained inside it to transfer.

Babies are tremendous learning machines, and key advances for AI may soon be found in the architecture of their little brains. If you think an artificial intelligence model running on thousands of cutting-edge computer chips is smart, allow me to introduce you to the concept of a 1-year-old. OK, so babies might not be able to write computer programs, solve advanced math problems, or debate philosophical ideas. But unlike today's AI models, which consume an ocean's worth of training data and as much energy as a small country, babies learn to make sense of the world with amazing efficiency. They identify new objects after seeing them once or twice, and they learn through fleeting observation and physical interaction. When it comes to improving AI, babies--and the architecture of their brains--might hold crucial insights.

Google is rolling out a search services history setting that controls whether Google Lens images, voice search audio and AI mode interactions are saved to train AI.

From the 1st-5th June, the robots descended on Vienna. The 2026 IEEE International Conference on Robotics & Automation (ICRA) brought together the top minds in robotics for one short week to showcase the latest technologies, form new collaborations, and exchange ideas. Held at the Messe Wien, a stone's throw from the bank of the Danube, ICRA proved to be equal parts technological marvel and thought-provoking discussion. The host venue for ICRA 2026: Messe Wien, also known as VIECON. My week at ICRA began with the 2nd ICRA 2026 Workshop on Robot Ethics: Ethical, Legal and User Perspectives in Robotics & Automation (WOROBET) .

Large vision-language models (LVLMs) excel at multimodal understanding but suffer from high computational costs due to redundant vision tokens. Existing pruning methods typically rely on single-layer attention scores to rank and prune redundant visual tokens to solve this inefficiency. However, as the interaction between tokens and layers is complicated, this raises a basic question: Is such a simple single-layer criterion sufficient to identify redundancy? To answer this question, we rethink the emergence of redundant visual tokens from a fundamental perspective: information flow, which models the interaction between tokens and layers by capturing how information moves between tokens across layers. We find (1) the CLS token acts as an information relay, which can simplify the complicated flow analysis; (2) the redundancy emerges progressively and dynamically via layer-wise attention concentration; and (3) relying solely on attention scores from single layers can lead to contradictory redundancy identification. Based on this, we propose FlowCut, an information-flow-aware pruning framework, mitigating the insufficiency of the current criterion for identifying redundant tokens and better aligning with the model's inherent behaviors. Extensive experiments show FlowCut achieves superior results, outperforming SoTA by 1.6% on LLaVA-1.5-7B with 88.9% token reduction, and by 4.3% on LLaVA-NeXT-7B with 94.4% reduction, delivering 3.2$\times$ speed-up in the prefilling stage.

Safety classifiers deployed in production operate under a stationarity assumption that fails silently: when input distributions drift, accuracy degrades with no error signal until ground-truth labels arrive. We present an online monitor that detects distributional shift in classifier scores via a sliding-window KS statistic with empirically calibrated alarm thresholds. In a pre-registered factorial evaluation (4 classifiers $\times$ 5 shift conditions $\times$ 20 seeds $\times$ 2 window sizes; 800 cells), the monitor achieves 86.6% valid detection (mean latency 39.5 steps) across synthetic-onset, real-jailbreak, and adversarial regimes; a classifier $\times$ shift interaction ($η^2 = 0.185$) shows that monitoring must be tuned per classifier. Attempting to recover post-detection coverage via weighted conformal prediction exposes a failure mode: density-ratio estimation collapses for generative classifiers because logistic regression separates source from target perfectly in 3584-4096-dimensional embedding space, clipping all importance weights to zero; projecting to $\leq 32$ dimensions restores coverage. We then extend the framework to gradient-based evasion and give the first threat-model characterisation of score-disagreement monitoring as a canary. We falsify three assumptions: that architectural diversity drives the signal (false, $η^2 = 0.011$), that it is generic out-of-distribution detection (false, GCG-specific, $p < 10^{-12}$), and that an adaptive attacker can suppress it (false while the canary is confident). We derive the exact security boundary, a confidence-gated equilibrium at which a monitor-aware attacker stalls at gap $= 1/(2λ)$, and provide a calibration-free scan martingale achieving false-alarm rate $\leq 1\%$ across all classifiers with no per-model tuning.

NFL great Chris Johnson reveals common first symptom of paralyzing ALS that's left him unable to speak or move: 'He thought it was a pinched nerve' Why Thylane Blondeau renounced her'most beautiful girl in the world' title after struggling with childhood honour - as she prepares to take on new name after marrying French DJ Ben Attal Stay-alert warnings issued as sharks return to one of America's busiest beaches I thought my drinking was harmless until I realized I couldn't go a day without it. Then I discovered a $3 miracle pill that killed all my alcohol cravings... I'm completely cured I thought I knew the secret to great sex... then one man proved me so wrong: JANA HOCKING is mind-blown by trick that women over 40 are loving How well do you REALLY know America? Take our ultimate history quiz to find out... Prince Harry vows to'explore every option to bring Archie and Lilibet to the UK' after hitting out at'bizarre decision' to deny armed security At 45 I was plagued by muscle pain, brain fog and memory loss... but it wasn't the menopause. I caught a disease while sitting on my sofa. I escaped Europe's dying democracies and discovered the real secret to America's enduring success: AYAAN HIRSI ALI Physical therapist who killed newborn and threw her in dumpster was finally snared 17 years later thanks to COSTCO receipt... as she's given incredibly light sentence for manslaughter I feel terrible saying my husband's penis isn't big enough for me.