@KGeN_IO

KGeN collects data from verified humans in the Global South to train Physical AI and LLMs Backed by @AccelIndia @Prosus_Ventures @jump_

coinmarketcap.com/currencies/kgen/
Joined February 2023
  • Tweets4.5K
  • Following34
  • Followers21K
1.9KMedia
$3,400,000+ earned by our AI contributors. In just 2 months. To put that in perspective, it's what some of the biggest chains in crypto have disclosed as their revenue within the same time span. The difference is this went straight to people. Our verified contributors are spread across 30+ countries, training AI across 15+ environments and 300+ task types. Robots learn from humans and those humans deserve the credit and the earnings. Like ours do.
34
14
1
61
47,983
Multiple devices. One job: teaching Physical AI how humans see the world.
4
5
18
1,564
What’s one daily task you do that looks simple but would actually break a robot?
7
4
27
3,000
Great breakdown from @humynlabs. As robots move into homes and workplaces, what matters is whether real people recorded their training data in the kinds of environments where these robots will actually operate.
Humyn Labs
@humynlabsOct 22
Text models learned from the internet. Robots have to learn from the real world. Joint trajectories, depth, tactile feedback, all in sync. We break down what physical AI training data looks like, and why diversity matters as much as volume.

What Training Data Does Physical AI Need?

Article · @humynlabs
5
6
32
2,668
Before Physical AI learns from a clip, we make sure a real human recorded it. That's our verified human network.
4
7
23
2,781
Bad data makes a chatbot say the wrong thing, but it makes a robot do the wrong thing. That's why AI safety starts at the source, with the data it learns from.

AI's Biggest Risk Starts at the Source

Article · @KGeN_IO
6
5
27
2,759
$71M+ base:0xf3d5b4c34ed623478cc5141861776e6cf7ae3a1e Volume in 7 days.
11
9
1
50
4,476
We've already seen flying cars, are we close to a future where robots will have their own calendars and routines?
15
7
58
90,697
In just 7 days, our top 1,000 contributors collectively earned $100K+ by recording real world data to help train Physical AI. As AI learns to understand the world we live in, the people capturing everyday life are earning from the work they put in to make it happen.
17
11
2
60
60,471
KGeN Foundation 🟩 retweeted
OrangeX
@OrangeXExchangeOct 12
Deposit, trade, and share a massive 143,000 base:0xf3d5b4c34ed623478cc5141861776e6cf7ae3a1e prize pool on OrangeX now! 🍊 📅 Oct 13, 06:00 – Oct 20, 06:00 (UTC) 🔹 Twitter Exclusive Event Stand a chance to win 200USDT within 3 steps: 1⃣ Follow @OrangeXExchange and @KGeN_IO 2⃣ Like and RT this post 3⃣ mention 3 friends in the comments 20 lucky users will be selected and share the 200USDT perpetual coupon 🔹 Trade KGEN Perpetual Trade over $1,000 to rank up and claim your share of 100,000 base:0xf3d5b4c34ed623478cc5141861776e6cf7ae3a1e (Top 1 wins 20,000 base:0xf3d5b4c34ed623478cc5141861776e6cf7ae3a1e!)🚀 🔹 Deposit Rewards Net deposit at least 1,000 KGEN to share a 43,000 base:0xf3d5b4c34ed623478cc5141861776e6cf7ae3a1e pool! 👇 Jump into the action and start trading: orangex.com/support/2/articl…
85
97
12
130
18,531
Context Matters. Even a simple task like folding clothes changes with the environment. Indoor or outdoor, lighting, shadows and space all affect how the task is performed. For Physical AI to get it right, it needs to learn everyday tasks across different environments and conditions. That’s why diverse, real world egocentric data isn’t optional. It’s crucial.
19
9
2
42
72,408
base:0xf3d5b4c34ed623478cc5141861776e6cf7ae3a1e is live on Bithumb. A big week just got bigger. Here's a quick recap of everything that happened 🧵
빗썸 Bithumb@BithumbOfficialOct 12
📢 New Listing 🚀 케이젠(#KGEN) 원화 마켓 추가 안내 🚀 $KGEN/KRW will be listed on #Bithumb! 🔸 Details : feed.bithumb.com/notice/1655… #Bithumb #KGEN @KGeN_IO
38
24
4
85
8,794
more replies
Here's everywhere you can find base:0xf3d5b4c34ed623478cc5141861776e6cf7ae3a1e👇
2
9
465
2
9
391
82M Network Strength. $3.4B $KGEN volume. 3.4B user attributes. That’s September ‘26.
18
12
1
45
5,162
Another KBW wrapped up. Great panels and a lot of talk about the future of robotics. Seoul, until next time 🇰🇷
12
12
1
46
5,885
This is where Physical AI starts 🇮🇳
13
8
40
4,562
KGeN Foundation 🟩 retweeted
thehype.
@thehypedotnewsOct 1
your home robot is dangerous, if it learned from the wrong data imagine your robot at home hitting cups like the karate kid instead of picking them up – and that can happen just because of a 1 mm error inside the robot training data. unlike with llms, the consequences are huge. @ishank20 sees this every day: his team records the hands robots learn from. full clip below 👇
thehype.
@thehypedotnewsSep 29
your next friend will be a robot to get there, it has to see, hear, touch and move like us. @AddyCrezee sat down with @manishdiesel and @ishank20, co-founders of @humynlabs – the team that has filmed 1m+ first-person videos and recorded 50k+ hours of audio of how people do it. they talked about the robot age – in homes, in factories and in how we talk to machines. • "robots as companions will be ahead of general purpose humanoids that will come and do your laundry" • "car covers one use case while robots are multiple use cases" – why robots will outnumber cars • "very few people in the world would be using keyboard and mouse to communicate with machines. it's all going to be voice" • "more robotic productivity than human productivity in as early as 2035" – quoting elon musk watch full interview below
3
3
5
12
6,124
Load more