标签：#bot

FCC ban on foreign-made robots accelerated RoboStore’s US manufacturing plans.

The Poolease X1 can collect leaves from the pool floor, but dirt, silt, walls, and steps are all beyond its pay grade.

The next-generation robotaxi, called the Waymo Ojai, is central to the company's push towards mass scale, and eventually, profitability.

During a recent visit to Generalist AI, I watched a robotic arm improvise and use a banana as a tool.

Robot vacuums are great at keeping your floors clean, but there are a few areas they fall down on the job - stairs being one. Another is chairs and stools. You know the scenario: You have a row of stools at a kitchen table, or chairs tucked under a dining table, and the robovac just […]

“We're not necessarily building in a dogmatic fashion towards full autonomy.”

The greatest invention in pet tech in recent years is the litter robot. A machine that scoops your kitties' poop so you don't have to - what else could a cat owner possibly want? How about insights into your kitty's litter box usage that could flag health issues? Sign me up. But I have two […]

When Claude Opus 5 and GPT-5.6 arrived, I expected my coding agents to become noticeably better. Instead, some of my workflows felt worse. The agents seemed more eager, less predictable, and occasionally “dumber” than before. Naturally, I blamed the new models. Very scientific of me. Maybe it was a skill issue Then I watched Andrej Karpathy’s interview on the No Priors podcast. One idea stuck with me: when an agent fails, the capability may already exist. The problem could be how we instruct it, what memory we provide, or how we arrange the workflow. That made me question something I had mostly ignored: Were my custom skills still compatible with the newer models? I gave my agent this prompt: Can you audit our custom skills against the current models? Flag stale prompts, conflicting instructions, outdated assumptions, and anything that should be simplified or removed. Then test each skill on a representative task and propose the smallest updates needed. The audit found instructions written around the behavior of older models. Some were redundant. Some were no longer necessary. Others pushed the newer models too hard and caused them to overdo tasks. After cleaning those up and testing the skills again, the results felt noticeably better. The official guidance supports this This was not proof that every disappointing result is caused by an outdated prompt. Models can still regress, behave inconsistently, or introduce genuine breaking changes. But both Anthropic and OpenAI recommend recalibrating instructions during model migrations. Anthropic’s Claude Opus 5 documentation says the model now verifies its work without being told. It specifically recommends removing verification instructions carried over from earlier models because they can cause over-verification. OpenAI’s GPT-5.6 guidance recommends removing repeated instructions, simplifying tool descriptions, and running the same evaluations after each change. In OpenAI’s internal coding-agent evaluations, leaner sys

Mastering Turn-Taking in Group Chat: How Two Characters Share One Thread Building a seamless group chat experience where two characters share a single thread can be surprisingly tricky. While one-on-one conversations with AI are relatively straightforward, introducing a second AI persona into the same chat thread raises a fundamental question: when a user speaks, who answers? We recently launched multi-character rooms on AmorLink, and this article delves into the turn-taking logic we developed. You'll discover why the majority of this logic deliberately avoids calling a language model and explore the contextual challenges that proved more complex than the routing itself. The Pitfalls of Simple Solutions The most intuitive approach is to have "everyone answer every message." However, this quickly devolves into a "press conference" scenario. Imagine asking, "How was your day?" and receiving two stacked paragraphs, each completely unaware of the other. This method also doubles inference costs and increases the time-to-first-token for every turn. Another common, yet flawed, strategy is to "pick at random." While cheaper, it's often more frustrating. If a user asks, "Iris, what do you think?" and the other character answers, the illusion of intelligent conversation shatters instantly. Randomness offers no improvement as the conversation scales. A Ladder Approach to Turn-Taking The key insight is that turn-taking isn't a single problem but a stack of them. The vast majority of these problems have unambiguous solutions. For instance, if a message explicitly names a character or is very short and follows a reply, the decision is clear. Only a minority of turns genuinely require complex judgment. Therefore, our policy is structured as a ladder, prioritizing cost-effective solutions for easy cases and reserving the more expensive AI model for the difficult ones. Our five-rung ladder works as follows, with the first matching condition determining the response: Exactly one memb

The Unitree G1 has found online fame as a relatively affordable robot that can charm a crowd. But can it ever hold down a real job?

Joi AI hired 10 people to masturbate using AI companions as part of a monthlong “wellness” study. The company claims the practice could help “solve male loneliness.”

The partnership is expanding beyond the initial market of Zagreb, Croatia to four additional European cities.

Human-AI marriages are not currently recognized by US law. Some Republican state policymakers are drafting legislation to keep it that way.

Robot lawnmowers are finally good enough to take a lot of work out of maintaining a yard, but they’re still not set-it-and-forget-it machines. If you don’t want these autonomous cutting machines to tear up your lawn or go roaming in your neighbors’ yard, you’re still going to need to keep an eye on them. But […]

Matic, my current favorite robot vacuum, just got a big upgrade. The company has launched Matic Cues, which brings voice and gesture control to the robot. Now, you can talk directly to your vacuum to tell it what you want it to do, or just point at a mess to have it spot-clean. The feature […]

Bartesian's cocktail makers would be best described like a Keurig or Nespresso machine, but for alcoholic drinks.

The Mission Robotic Vehicle is making the first attempt to attach a new thruster to an aging satellite.

Short answer: the least risky alternative to a single-provider OpenAI-compatible API is a thin routing layer with one internal contract, a small Python adapter, and an eval set that measures answer quality before price. Treat “cheapest” as a workload result, not a label. A US/EU chatbot also needs a deliberate data-residency decision before a key or SDK enters production. The attractive story is easy: one API key, one SDK, and a familiar chat-completions shape. Measure it. The production story has more edges. Provider-specific tool calls, token accounting, streaming events, retention settings, and regional routing can differ while the first text response still looks fine. That is how an in-app chatbot passes a demo and fails an eval. Consider a support bot that retrieves three passages, answers in a stream, and offers an escalation tool. A compatibility test that checks only the final sentence can miss an empty retrieval marker, a tool argument that is valid text but invalid JSON, a stream terminator that the client never handles, and a fallback that sends the same user request to a second region. The transcript still looks plausible in a screenshot. The trace tells a different story. I've learned to make those states explicit in the adapter before tuning a model. I build RAG and agent features in Python, so my first question is not “which model wins?” It is “which contract can I test?” The app should own that contract. A provider adapter should translate it at the boundary, and the rest of the application should never know whether the request went to an OpenAI-compatible endpoint, a Claude-style API, a Gemini-style API, or a local service. How can an app chatbot compare compatible API alternatives across US and EU? Start with the request that matters to the user: a message plus retrieved context, a latency budget, a maximum output, and a trace ID. Record the selected region and provider in server-side metadata, but don't send a secret to the browser. “One API key”

The divestiture comes as the two once-tight companies have started to diverge on the business side.

"Smile because it happened" — Dr. Seuss The Setup Earlier this year, I took on a short-term trial role with an early-stage robotics startup. The premise was straightforward: help with data collection, annotation, and evaluation workflows—essentially the backbone of any modern robotics or embodied-AI system. The trial didn't work out long-term. I was let go after about two months — a decision that, honestly, came down in part to my bandwidth as a student. Balancing a full course load with a startup trial was harder than I anticipated. But that's not the story I want to tell. What I do want to share are the technical lessons I took away — lessons about building robust data pipelines, about the gap between theory and practice, and about what I'd do differently next time. These aren't company secrets. They're about the general engineering challenges that anyone working with robotics data pipelines will encounter — challenges I'd read about in papers but hadn't truly internalized until I was standing in front of them. 1. The Data Pipeline Shape Is Universal—But the Details Aren't If you've spent any time in ML or robotics, you've seen this described: Data Collection → Annotation → Evaluation It's a standard three-stage pipeline. Industry vendors describe it explicitly in their robotics content. Academic projects model this structure. It's the field's shared vocabulary. Companies such as Scale AI and Toloka use similar industry workflows involving data collection, annotation, and evaluation. What isn't shared are the specifics: the sensor setup, the calibration procedures, the annotation rubric, and the evaluation metrics. Those are where a company's IP lives. The pipeline shape? That's just the map. And the map is public. What I'd do differently: Simulate before you collect. Data collection is expensive — in time, hardware wear, and cognitive load on operators. Before running a full session, run a feasibility study with a small batch. Verify your sync and capture scripts