Physical AI is entering a new phase where real world data and community. contributions could become just as important as models and compute. And This is one reason I’ve been paying attention to @axisrobotics I like the idea of connecting everyday contributors with the development of Robotics AI, rather than keeping everything limited to labs and engineers. Because smarter models alone aren’t enough. Robots need to learn from real-world movements, human feedback, and different scenarios to better understand how to operate outside controlled environments. Axis is focused on the data layer behind robotics: • Quality Trajectories • human feedback to correct mistakes • diverse scenarios to improve how robots learn Physical AI is still in its early stages, and many challenges remain. But the projects building both technological foundations and sustainable communities deserve attention. If Physical AI becomes one of the next major AI narratives, the projects building around both data and community could be worth watching early.

Missed the @conso_xyz BOT eligibility list? Don't worry, there's still another way to get it You can earn up to 10 CONSO BOTs for free just by creating content about CONSO on X. Here's how it works: • Publish up to 3 posts about CONSO • Get impressions and turn your reach into BOT rewards The reward system : • 1K impressions → 1 BOT • 3K impressions → 2 BOTs • 5K impressions → 3 BOTs • Every additional 5K impressions after that earns you 1 more BOT The maximum is 10 BOTs per account. so if you didn’t make the eligibility list, you still have a chance to earn your way in. Time to put your posts to work.

October 21 could be a Big day for $AXIS CoinMarketCap is showing Oct 21 for the unlock date. Nothing official yet, but Waiting for @axisrobotics to confirm. Before Snapshot , Dont Forget Here's the main thing : • Trajectories sign-ups end Oct 18 • Kaito creator campaign ends the same day • $sKAITO registration also closes on Oct 18 Also, late Trajectories sign-ups will count toward S2, with a bigger allocation. Still time to get in if you haven't joined yet : s.kaito.ai/vFSEhob

Glad to see $PUMP narrative gets more interesting when you look at how much of its circulating supply is tied up in memecoins. people might laugh at this take now, but I can actually see $PUMP making it into the top 5 someday

Lately, I've been thinking a lot about how far robotics could go in the next few years. I've been following Axis since the early days. We've made videos, tutorials, and educational content to help more people understand the project, and it's been really cool to see how much things have grown. Since launching on Base On March , @axisrobotics Make Good Progress now : • 7.66M+ Trajectories • 235K+ users • 7.28M+ transactions Beyond the numbers, it's exciting to see how manypeople are coming together to help train the next generation of Robots You can contribute right from your browser, without needing a fancy lab or expensive equipment. Every contribution adds to the growing pool of data that could help Robots learn new skills. Plus, signed Trajectories receive onchain Data IDs, making those contributions traceable. We often talk about what robots might be able to do in the future, but there's a lot of work that goes into teaching them first. Being part of this community its been Really Nice and seeing how far Axis has come makes the whole experience even moremeaningful. There's still a long way to go, but watching the community grow and the project make progress has been exciting.

Axis has announced a snapshot for Oct 18 at 12:00 UTC. All Axis points go into one reward pool, including Binance Wallet and BitRobot Alliance. Post Snapshot contributions still count, just toward the 6% post-TGE community pool. If you’ve been doing tasks on @axisrobotics don’t forget to check your signatures , go to : • Axis Hub • Portfolio • History • and check if you have any completed trajectories that still need signing. You might’ve finished the tasks, but without signing on Base, those Trajectories won’t count toward your Epoch points You’ve already done the work, so don’t let an unfinished signature go to waste. Better , take a minute and check your account before the snapshot

Simulation can teach a robot what to do, but it doesn’t guarantee the skill will translate outside the Sim. @axisrobotics is Working on a problem that goes beyond just training Robots. The goal is to turn Skills learned in simulation into policies that can adapt to different object positions using visual input. if that transfer works reliably, it could make robot training a lot more scalable. because a Robot looking good in a controlled environment is one thing. Making it work in the real world is a different challenge.

4 days left and you’re not late to join Axis Katalyst Epoch 2 on Kaito. • With 2.5M $AXIS total • 2M for the top 500 creators • 500K for Kaito stakers • fully unlocked at TGE if you’re joining the @axisrobotics campaign, don’t just focus on posting. Make Good content, get your Axis Hub tasks done, and invite more Friend To Join. stay active and make your contributions count 👉 s.kaito.ai/vFSEhob

Miden mainnet is almost here Privacy and scalability are a big part of what makes miden interesting. let's see what the team has been cooking. bullish to see what's next.

One thing I found interesting in @axisrobotics Latest research is how much the choice of data matters. More data doesn’t always mean better results in Robotics. Sometimes, it’s about choosing the right data. In Axis study, 600 carefully selected simulation Demos Achieved 85.8% average success across three manipulation tasks. For comparison, using all 1,200 Demos or 600 randomly selected ones only reached 25.8%. For setup was consistent. all three tasks shared the same 10 successful real world demonstrations. Each condition was evaluated across 40 trials per task. TUCO takes a smarter approach, selecting demosthat help the robot learn more effectively. But there’s still a catch. When the lighting changed, peg success dropped from 82.5% to 25%. That’s a pretty big gap, and a reminder that doing well under controlled conditions doesn’t automatically mean a robot is ready for the real world. Picking the right data is promising. Proving it works outside the lab is the next step. I’d love to see how this holds up with changing lighting, unfamiliar objects, and less predictable environments. Want to see how the data layer of Physical AI works in the real world? Join $AXIS and try the system yourself, and explore how Physical world data is being generated. 👉 s.kaito.ai/vFSEhob

Yesterday Axis Tokenomics Are Out As someone who’s been following @axisrobotics , I wanted to look beyond the numbers and understand how $AXIS fits into the bigger picture. Let’s start with the tokenomics. $AXIS has a fixed supply of 1 billion tokens Here’s the breakdown: • 61.08% Ecosystem & contributor rewards
• 31.67% Team & investors
• 6% Liquidity
• 1.25% Community sale & campaigns At TGE, 16.5% of the total supply will be unlocked. 100M $AXIS, or 10% of the total supply, is reserved for contributors across two seasons: → Season 1: 40M $AXIS
→ Season 2: 60M $AXIS For Season 1, 50% unlocks at TGE, with the rest distributed at T+3 and T+6. If AXIS reaches a $250M FDV, The potential rewards for the Top 100 contributors could be around $2.5K per person on average, depending on the actual reward allocation, vesting schedule, and token price. Of course, this is only a potential valuation scenario, not a guaranteed return Why does contributor participation matter in the first place? Because Physical AI needs real world data to help robots understand their environments, learn from experience, and perform tasks more effectively. $AXIS aims to drive this progress by incentivizing contributors and supporting the growth of the ecosystem. And With the progress so far is worth watching : • 6M+ trajectories in September, with 100K+ generated daily • 3M+ trajectories recorded • $12M seed round led by Hack VC • 1.5M Axis Points Binance Wallet campaign • $AXIS Community Sale oversubscribed These milestones provide a glimpse into the ecosystem Axis is building and the role real world data could play in advancing Robotics. Tokenomics can create incentives, but long term success will ultimately depend on execution, adoption, and the value the ecosystem delivers. The next phase of Axis is definitely worth watching.

AI makes personal brands more powerful than ever. One person can now build, influence, and move resources at scale. But when everyone can create anything, reputation becomes the real currency. Power gets attention. Trust sustains it.

September wil be biggest Month For @axisrobotics More data flowing on $BASE with : • 6M+ robot Trajectories • 160K+ downloads for the Franka dataset on HuggingFace At the same time, the Open Axis Benchmark Launched with over 6,000 tasks for the robotics ecosystem. Community momentum was just as strong. The Community Sale raised $2.39M from 2,472 participants, making it Sonar’s most participated sale of 2026. Meanwhile, CEO Chris shared perspectives with Brian Armstrong on the future of Physical AI , and Axis research reached the IROS 2026 PWMS workshop and pushes Embodied RSI forward with success rates climbing from 22% to 52% Beyond the core updates, Community Momentum is still going strong with : The community engine kept moving too: • 2,521 contributors on BinanceWallet campaign • 477,461 task completions • 989 tasks covered • Kaito Content Creator Program entered Epoch 2 September was more than just a month of updates for Axis. We got to see the work behind the vision, from research and infrastructure to real community participation. Because building Physical not just about making better models, But giving them more real world experience to learn from and improve over time. Now Let’s see what October brings For Axis

Axis TGE might be closer than we think . The more I look into @axisrobotics the more I believe Physical AI needs more than just powerful models. Robots need quality data, real world experience, and a way to measure whether they can actually get the job done. That’s what makes Axis interesting. From collecting robot trajectories to verifying data, training models, and testing performance through Open Axis Benchmark, they’re building a full loop around improving how robots learn. I’ve been contributing through the Kaito Axis Content Creator Program, while keeping an eye on how this ecosystem develops on @base But raw data numbers aren’t everything. What really matters is whether that data helps robots handle new tasks, perform more reliably, and move from simulation into the real world. If Axis can turn community contributions into better models and real commercial use, this could become much bigger than a robot data platform. Still watching how this plays out. The real test is what gets built from all that data. Explore the Axis workflow and find your own way to contribute. Join Here : s.kaito.ai/vFSEhob

Been keeping an eye on this one @FungoLabs is building what they describe as the first encrypted NFT collection on Ethereum mainnet, powered by @zama 3,232 NFTs total:
→ 500 free mint
→ 232 team
→ 2,500 public sale So if you’re curious, there’s still time to apply for the free WL: 
fungolabs.org/apply I’ll admit I’m probably a bit biased because anything built around Zama gets my attention . Still, DYOR before touching the public sale. NFA.

honestly @arc is still under the radar every time i look around there’s another project i hadn’t heard of kinda fun finding them before everyone starts talking about it

Big Day For Jumper Community Finally $JUMP Is Coming If the airdrop rumors are right, things are about to get interesting

I like Solana . Fast, chaotic, and full of interesting experiments. And if you’re already exploring the Solana meme ecosystem, $KNOTS is one to keep on the radar Built on Solana, with a pretty interesting connection to $STONK $KNOTS

The $STONK and solana:8RVBk8vxLiUHueLUW1f4izFVqN3nWippLhkohKg6EGkS connection is actually pretty interesting. the real alpha is in how the two are connected. $KNOTS = activity $STONK = connected to the rewards side of that activity for qualifying holders so yeah, different tickers, but they’re actually built to work together. definitely more interesting once you look past the wordplay.