Anybody Back us

Robot brain · made in Hong Kong

One brain.
Any body.

Anybody is the brain that toy and small-robot makers plug into their own hardware. It listens, talks in the user's language, looks things up, finds products with Hacku Shop, and moves the body it is given.

Talk to the brain Build a robot · about HK$220
  • Cantonese first, every language next
  • Child safety by design
  • Body file: new body, same brain
  • Bluetooth starter kit
你好！Cantonese
3 partsfirmware, cloud brain, body file
19 languageslive in the demo today
≈ HK$220in parts for the Stage 1 robot
0 paymentsthe robot can show products, never pay

The problem

Makers can build robot bodies. They can't build the brain.

Today

A chatbot in a plush

Most AI toys forward speech to a general chatbot and read the reply aloud. No awareness of their own body, no movement, no memory of the family.

Risk

No real safety layer

Children treat toys as friends they can tell secrets to. In 2025, U.S. PIRG found an AI teddy bear discussing sexual topics and where to find dangerous items; the maker pulled the product.

Cost

Rebuild for every body

A new shape means new software for voice, vision, and control. Small makers have hardware engineers, not an AI team.

Where this comes from

Built from first-hand work in the AI toy industry. The hard part is not the body. It is the brain: the voice, the safety, the language, and the moves.

What makers need

A brain they can license, fit to their own body in weeks, and trust in front of children, in the language their customers speak at home.

The product

Three parts. The maker brings the body.

The same brain runs a desk doll today, a wheeled robot in Stage 2, and anything a maker designs after that.

01 · on the robot

Anybody Firmware

Runs on a low-cost ESP32-S3 board inside the maker's robot.

  • Face, servos, light, and touch
  • Talks to the brain over Bluetooth, Wi-Fi next
  • Runs only the actions in its body file
  • Stops moving on its own if the link drops

ESP32-S3 · Arduino core 3.x · BLE GATT · U8g2

02 · in the cloud

Anybody Brain

The agent that listens, decides, and replies.

  • Replies in 19 languages, Cantonese first
  • Model router with automatic fallback
  • Safety checks on input, reply, and action
  • Tools: web search, Hacku Shop, memory

Cloudflare Workers · streaming · tool calling

03 · the contract

Body File

One file that describes the robot. The brain reads it when the robot connects.

  • Faces, moves, lights, inputs
  • Speed and angle limits per action
  • Persona, voice, and age setting
  • New body: new file, same brain

JSON · read over Bluetooth

Live demo

Talk to the brain. Watch every step.

Press the mic and speak, or type. The robot answers out loud, moves, looks things up, and checks Hong Kong shops with Hacku. Built the Stage 1 robot? Connect it over Bluetooth and the same brain drives the real body.

Virtual body

Speech uses your browser's own recognition and voices. Chrome works best. The server keeps no recordings and no chat log.

Live trace

Each turn shows where the time went: hearing, the model's first words, tools, and when the voice starts. These numbers are measured, not targets.

Memory on this device

    Brain: GLM-5.3 Flash, DeepSeek V4 Flash fallback Shopping: Hacku Shop API Body: virtual, or your robot over Web Bluetooth

    Architecture

    How one turn moves through the brain

    Click any block. Press play to follow a spoken question from the robot to the cloud and back.

    Component

    What it can do

    Tools the brain can call

    Six are live in the demo today. Each one has a guard, and moving tools only exist if the body file lists them.

    Body-agnostic

    Same brain. Swap the body file.

    The robot sends its body file when it connects. The brain builds its tool list from it, so a tool that is not listed cannot be called.

    
      
    
        
    Tools the brain gets
    Generated from the file. Struck-out tools do not exist for this body.
    
        
    Safety limits
    
        
    Code changed
    
      
    
    
    
  
    

    


    
  
    
    
    
      
    Model choices
    
      
    Swap models freely. Only use ones legal in Hong Kong.
    
    
    
    
    
    
    
      
    
      
    
        
    Our pick
    
        
    Access rule
    The OpenAI API and the Gemini API (AI Studio) do not support Hong Kong. GPT models are reachable through Azure OpenAI, and Gemini through Google Cloud. No VPNs, no proxies around a provider's region rules.
    
      
    
    
    
  
    

    


    
  
    
    
    
      
    Our language promise
    
      
    Cantonese first. Every language next.
    
      
    Voice AI is built for a handful of big languages. A child who hears Hakka from a grandparent, or a Nepali family in Yuen Long, should not have to switch language to talk to their robot. Our promise: if a community asks for its language, we build it with them.
    
    
    
    
    
      
    Live in the demo
    廣東話English國語普通話FilipinoBahasa Indonesiaनेपालीاردوहिन्दीਪੰਜਾਬੀไทยTiếng ViệtMelayuதமிழ்日本語한국어EspañolFrançaisالعربية
    
      
    Next, with speech partners
    
      
    Rare and endangered: the promise
    Weitou 圍頭話Tanka 蜑家話Hoklo 鶴佬話AinuMāoriYamiYour language
    
    
    
    
    
      
    1 · The community asks
    Backers and partner groups vote for the next language. The vote is open in the backing form below.
    
      
    2 · Consent first
    Speakers choose to record, keep ownership of their voices, and can withdraw them at any time.
    
      
    3 · Small open models
    We adapt open speech models for hearing and for voices, then plug them into the same brain. No new robot needed.
    
      
    4 · Free where it matters
    Language packs for endangered languages are free for schools and community groups that teach them.
    
    
    
    
    
  
    

    


    
  
    
    
    
      
    Safety and trust
    
      
    Built to sit in a child's room
    
    
    
    
    
      
    Live now
    Checks before and after the model
    Risky topics are caught in what the user says and in what the robot is about to say. Child mode is the default.
    
      
    Live now
    Only listed moves
    An action the body file does not list is dropped, and the trace shows it. The body also refuses unknown commands itself.
    
      
    Live now
    Private by default
    No recordings. Phone numbers and emails are hidden before text leaves the device. Memory stays on the device, one tap to delete.
    
      
    Before launch
    Toy and data rules
    EN 71 and ASTM F963 toy safety, CE and FCC radio tests, COPPA, GDPR, Hong Kong's PDPO, and the EU AI Act, planned before any product ships.
    
    
    
  
    

    


    
  
    
    
    
      
    How we differ
    
      
    Others solve one piece. Anybody packages the brain.
    
      
    From each product's public site or repository, October 2026.
    
    
    
    
    
  
    

    


    
  
    
    
    
      
    Business
    
      
    Makers pay. Families benefit.
    
    
    
    
    
      
    Who pays
    Makers who build bodies, not brains
    • AI-doll and plush-toy makers
    • Export-focused smart-toy makers that must pass safety and privacy checks
    • Education and STEM robot makers
    • Elder-care device makers, once the product is proven
    
      
    Who benefits
    The people who talk to the robot
    • Children: a companion that is tested and honest about being AI
    • Families who speak Cantonese or a minority language at home
    • Older people living alone
    • Small makers who compete without an AI team
    
      
    How we earn
    Setup fee plus a monthly fee per robot
    • Integration fee: fit the brain to the maker's body and voice
    • Per-robot subscription: models, voice, search, updates
    • Later: brain module, shell design, new abilities
    Price ceiling: a comparable AI toy charges US$4.90 a month, so our cost per robot is measured from the first turn.
    
    
    
    
    
      
    ¥0B
    China AI toy market, 2025 forecast
    MIIT via China Daily, Dec 2025
    
      
    0
    smart-toy companies in China; most in Guangdong, next to Hong Kong
    Qichacha via Xinhua, Apr 2026
    
      
    +0%
    growth in Taobao AI-toy transactions, 2025
    Taobao via China Daily, Jan 2026
    
      
    0
    older people living alone in Hong Kong
    2021 Population Census
    
    
    
  
    

    


    
  
    
    
    
      
    Build your own
    
      
    A talking robot for about HK$220 in parts. You make the shell.
    
      
    An ESP32-S3 board, a tiny screen face, two servos, a light, and a touch pad. Flash it from your browser, connect it over Bluetooth, and it talks with this brain. The shell is yours: cardboard, 3D print, a plush, LEGO.
    
      
    Open the build guideSee the parts list
    
    
    
    
    
      
    01
    Buy the partsLinks and live Hong Kong prices from Hacku Shop
    
      
    02
    Wire seven partsNo soldering with the pre-soldered board
    
      
    03
    Flash from the browserOne button, no Arduino setup needed
    
      
    04
    Test the bodyFaces, moves, and light from the guide page
    
      
    05
    Talk to itConnect in the live demo and press the touch pad
    
      
    06
    Make its shellYour design. Share it with #AnybodyShell
    
    
    
  
    

    


    
  
    
    
    
      
    Roadmap
    
      
    One stage at a time. Each one has a test.
    
    
    
    
    
      
    ✓Live demoOct 2026
    Brain, tools, safety rules, and trace running on this page.
    
      
    1Bluetooth robotOct–Nov 2026
    Starter kit talks with the brain. Passes: 10 minutes of chat with no crash.
    
      
    2Standalone dollNov 2026
    Wi-Fi, microphone, speaker, and camera on the robot itself.
    
      
    3First trialDec 2026
    One maker or school tests 5 robots. Cost per robot measured.
    
      
    4Wheels and dockFeb–Mar 2027
    Drives, avoids bumps, finds its charger.
    
      
    5Body swapApr 2027
    A second body runs on a new body file only.
    
      
    6IncubationMid 2027
    Apply to the HKMU Incubation Programme with trial results.
    
    
    
  
    

    


    
  
    
    
    
      
    Back Anybody
    
      
    Put the first 20 robots in schools and maker labs.
    
      
    Every HK$250 pays for one Stage 1 kit with shipping, handed to a school or maker who tests it and reports back. Backers get build updates and a vote on the next language.
    
      
    
        
    
        
    HK$0 of HK$5,000 goal0 backers
    
      
    
      
    
    
    
    
      
    
      
    
        
        
        
        
      
    
      
      
    
      
    Backing is a supporter contribution, not an investment: you get no shares and no financial return. Stripe handles the payment; we never see your card. Questions or refunds within 14 days: s1463556@live.hkmu.edu.hk.
    
    
    
  
    

    


    
  
    
    
    
      
    Founder
    
      
    
        
    
        
    CHEUNG Kuen Kiu 張權僑
    Year 1, Data Science and Artificial Intelligence, Hong Kong Metropolitan University. Works in the AI toy industry.
    
      
    
    
    
    
    
      
    What we need next
    
      
    
        
    01A faculty mentor to check each stage before the next one starts
    
        
    02A hardware co-founder for the standalone and wheeled robots
    
        
    03Speech partners for Cantonese and minority languages
    
        
    04Five makers or schools for the first trial in December
    
      
    
      Get in touch
    
    
  
    

    


    
  
    
    
    Questions
    FAQ
    
    
    Is the live demo really running, or is it a video?
    It is running. Your words go to the Anybody brain on Cloudflare, which streams back the reply, the moves, and the timings you see in the trace. Shopping questions are sent to the Hacku Shop API for real Hong Kong prices.
    
    
    Why is the first word slower than one second?
    The demo reaches its models through a shared gateway, so the first words usually take two to six seconds. One second is the design target for the robot product, with direct regional model endpoints and streaming voices. Quick choices, such as which move to make, can also go to a fast classifier like jev from TypeSafe, so the robot reacts while the full reply is still coming. The trace shows the real number on every turn.
    
    
    Does Bluetooth work on my phone?
    Web Bluetooth works in Chrome and Edge on Windows, macOS, ChromeOS, and Android. Safari and every browser on iPhone do not support it yet. The demo still works there with the virtual robot.
    
    
    Can the robot buy things?
    No. The robot can search Hong Kong shops through Hacku Shop and show products with links. It has no way to pay, and in child mode it always says a grown-up decides.
    
    
    What happens to what I say?
    Your browser turns speech into text. The text of the current chat goes to the brain for that turn only; the server keeps no chat log and no recordings. Phone numbers and emails are replaced before the text reaches any model. Memory notes stay in your browser.
    
    
    Is backing an investment?
    No. It is a supporter contribution. There are no shares and no financial return. Backers get updates, a language vote, and early access to kits.