--- title: "Gemini Robotics 2 brings whole body intelligence to robots" url: https://daily.dev/posts/gemini-robotics-2-brings-whole-body-intelligence-to-robots-1tocz8csn source_url: https://deepmind.google/blog/gemini-robotics-2-brings-whole-body-intelligence-to-robots type: article source: "DeepMind" published: 2026-07-30T15:09:04.043Z updated: 2026-08-01T20:18:20.072Z tags: ["robotics", "google-gemini", "google-deepmind"] reading_time: 10 upvotes: 1 comments: 0 language: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Gemini Robotics 2 brings whole body intelligence to robots **[DeepMind](https://daily.dev/sources/dm)** · 10 min read · 1 upvotes · 0 comments ## Summary Google DeepMind has introduced Gemini Robotics 2, a suite of three AI models designed to give robots intelligent whole-body control, advanced dexterity, and multi-robot collaboration capabilities. The suite includes: Gemini Robotics 2 (a vision-language-action model enabling full humanoid control from feet to fingertips), Gemini Robotics ER 2 (an embodied reasoning model for multi-step task planning and robot teamwork), and Gemini Robotics On-Device 2 (an efficient on-device VLA that can adapt to new robot embodiments in a few hours with under 200 examples). Key advances include controlling Apptronik's Apollo 2 humanoid for whole-body tasks, fine dexterous manipulation with 22-DOF hands, agentic reasoning for multi-minute task sequences, and a new ASIMOV-Agentic safety benchmark. The ER 2 model is available on Google AI Studio and Gemini Enterprise Agent Platform, while VLA models are in early-access partner preview. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Community take How the wider developer community reacted, aggregated from 1 discussion and 194 comments across hackernews (as of 2026-08-01). **TL;DR:** The community is cautiously optimistic about AI-powered humanoid robots but skeptical about near-term household viability, debating cost barriers, safety risks, and whether progress will mirror LLMs; business/industrial use cases are seen as more tractable than home deployment. **Sentiment:** 35% positive · 40% mixed · 25% skeptical **The case for** - Humanoid form factor is well-suited to the already human-built world, requiring no infrastructure changes. - Progress could mirror LLM scaling — early slowness doesn't preclude rapid future improvement. - Leasing/subscription models could lower the upfront cost barrier for consumers. - For households already paying for cleaning services, a robot could pay for itself over time. - Businesses have a clear ROI case: replacing multiple human shifts with a single capital purchase. **The pushback** - Current robots are slow, have low task success rates (e.g. ~36% on screwing in a lightbulb), and are far from reliable general-purpose assistants. - Safety is a serious concern — a large motorized robot near children or unsupervised could cause significant harm or property damage. - LLM-style malicious compliance and hallucination errors could be catastrophic in physical environments. - High upfront cost puts household robots out of reach for most consumers; even $20k is top-10% territory. - Dexterity remains a fundamental hardware problem — grippers aren't hands, and multi-finger manipulation is still very hard. - Maintenance complexity, software update unpredictability, and potential subscription lock-in add ongoing risk. **By community** - hackernews (mixed): Commenters are split between excitement about the LLM-like trajectory of robotics and deep skepticism about near-term household viability, cost, safety, and the gap between demos and real-world reliability. **Hottest debate:** Whether household humanoid robots will follow LLM-style rapid progress or remain stuck like self-driving cars — and whether the cost and safety barriers make consumer adoption realistic at all. **Open questions** - What are the actual inference costs per task, and when will they fall below the cost of human labor? - Will VLM/VLA architectures be sufficient for robust dexterity, or does LeCun's world-model approach have merit? - How will liability and safety regulation be handled for unsupervised humanoid robots in homes? - Who will service and maintain these robots, given their extreme mechanical complexity? - Will capital from humanoid robots be democratized, or will it further concentrate wealth? **Highlights** > I'm bearish that they'll be viable in the house for a very long time. For businesses the bar for adoption is very low: If the thing can work repetitive jobs for 24 hours a day and replace 3 shifts, the purchase bar is nominally anything less than 3 x human salary. That's a high number, and probably fairly easy to achieve. For homes, it's a very different bar. You'd have a hard time convincing most American families to purchase anything with a >$1000 price tag. > — [qurren on hackernews · 7 comments](https://news.ycombinator.com/item?id=49111728) > The time between gpt 2 and 3 was 15 months. The time between gemini robotics 1 and 2 was also 15 months: https://blog.google/products-and-platforms/products/gemini/h... The difference between gpt 2 and 3 was insane. 2 could generate limericks when it wasn't repeating a word 300x. 3 could actually do some things. By comparison gemini robotics has hardly changed at all. I will also point out that slow, non-fluid robotics is on a totally different level of difficulty from fast fluid motion. Asimov could walk pretty smoothly, but it didn't fall over because it used a very careful sequence that was never unbalanced; you could pause at any point without falling over. Move faster, like boston dynamics, and you need to account for the change in balance from your arms swinging... or rather, you need to be able to account for the rotational inertia etc from moving multiple masses along complex paths with multiple points of articulation at hundreds or thousands of times per second. An algorithm to fold tshirts 90% of the time is easy. The cloth hangs down by gravity and you can just look for right angles (corners), find their coordinates with binocular matching, and move them to meet each other. Getting 99%, or folding them quickly, so that the fabric is actually moving instead of just hanging still- incredibly, incredibly more complex. > — [hwillis on hackernews](https://news.ycombinator.com/item?id=49112035) > IMHO the real test is if any robotic startup currently selling (or planning to sell) robots as a service for homes not just use it but gets returning users from it. I did professionally few prototypes with robots and progress is real yet very far from what the average customer would find reliably useful in menial tasks. FWIW I do think https://rodneybrooks.com/why-todays-humanoids-wont-learn-dex... remains relevant, namely dexterity is also a hardware problem, grippers aren't hands. They even clarify "multi-finger dexterous manipulation remains challenging." and those aren't even fingers with a lot of sensors. > — [utopiah on hackernews](https://news.ycombinator.com/item?id=49111583) > "That's fair, you are absolutely right I should not have unscrewed the water hose and put it away before turning on the valve to water the plants, even though you specified that order, I should've used some common sense." There is going to be so much pain from VLA malicious compliance. If the current gen of LLMs are anything to go by I can already see it being an hilariously massive problem. > — [moffkalast on hackernews · 1 comments](https://news.ycombinator.com/item?id=49112512) > I work in this field, and I wrote my bachelor's thesis here, not with humanoids but with VLAs (think chatgpt connected to a robot arm) It's certainly not there yet for anything practical, there's also certain bits and structures that don't have accurate names during construction Plus we have no good reliable accuracy testing data in most cases (most tests occur on a few demos, but that isn't a good representation of how must things work), popular benchmarks, such as libero have been saturated, and nearly everything gets 95% there, most companies and researchers have their own benchmarks here. Plus companies lie alot, and do very dangerous things in thier videos, I.e. these robots should not be standing very close to humans, because of being dangerous. There are also legitimate concerns of misuse of these robots that need to be accounted for, misuse does not have to be warfare, but can be as simple as confusing it while it is cutting tomatoes with a knife. Turning doorknob is easy, and fail recovery is also being worked on, but we don't have reliable statistics anywhere on that. The hard part is on practical things, as in when placing bricks or attaching a part during manufacturing it needs to ensure that it is aligning everything correctly....and that's hard, while it is impressive, it is very irresponsible to keep humanoids at home (people are irresponsible when untrained), for example, lawnmowers injure about 6400 people a year...and that is not an everything machine. Humanoids in general are...not appealing in specific, due to maintainable of joints, complexity, but robot arms in particular, expecially on wheels (check mobile aloha), are likely to be able to do tasks such as clean up in hotels, after a guest had left, or replace some cooks in restaurants (if their work is consistent) > — [adityashankar on hackernews](https://news.ycombinator.com/item?id=49111770) **Source threads** - [hackernews](https://news.ycombinator.com/item?id=49111237) · 74 points · 194 comments --- Tags: [#robotics](https://daily.dev/tags/robotics), [#google-gemini](https://daily.dev/tags/google-gemini), [#google-deepmind](https://daily.dev/tags/google-deepmind) [View this post on daily.dev](https://daily.dev/posts/gemini-robotics-2-brings-whole-body-intelligence-to-robots-1tocz8csn)