AgentSuite is your portal to the agentic economy. Create your own agent on https://xitter.cf/t.co/3g42YjcIan. Built on VeChain.
Joined May 2026
- Tweets530
- Following21
- Followers767
- Likes308
Pinned Tweet
VeWorld Lite is now on the App Store.
Build your agents and talk to them straight from your phone.
Download it and give it a try.
Is AI in a bubble? And if it bursts, what happens next?
The AI investment boom may be due for a correction. But that doesn't mean AI itself is going away.
And as agents grow more capable and move from answering questions to getting things done, where will the lasting value actually come from?
In case you missed it, our latest Academy article revisits @sunshinelu24's perspective on what could survive the hype and shape the next phase of the AI economy.
Read the key takeaways now 👇
veworld.ai/academy/articles/…
New parents don't need another app that counts the wakings and shows a red chart. They need to know what's normal, and when it isn't.
Bloom explains calmly, and tells you when to call a doctor instead. Meet Bloom, a child development companion.
An agent gets useful when it stops needing instructions.
On @VeWorldAI you teach it your way of doing something once, give it your calendar and files to work with, and from there it runs on its own.
We're watching a race to figure out what comes after the touchscreen.
Voice assistants you have actual conversations with. Smart glasses. Tiny robots sitting on our desks. Even the apps we already use are changing. Different interfaces, same ambition: make interacting with AI feel as natural as reaching for your phone
The iPhone didn't invent touchscreens, but it helped make them part of everyday life. We will be witnessing something similar with AI.
Though maybe the next big breakthrough won't be a single device or interface at all. Talking to an agent while cooking, using a screen to review its work, or getting info through your glasses on the move. The future could be more about technology adapting to how we live.
Perhaps in five years, interacting with AI won't feel like a separate activity anymore.
Couldn't help but QT this. Kinda fascinating how much interface design can shape our relationship with AI.
A realistic avatar could make interacting with an agent feel more natural and enjoyable. Imagine learning something new or practicing a difficult conversation face-to-face with AI.
But while realism can make an interaction feel more human, trust should still come from what the agent actually does, how reliably it performs, and how accountable it is.
How do we make AI feel more human without making it seem more trustworthy than it really is?
After Google acquired my AR startup, I started thinking about the next interface, this time for AI. Text and voice work fine, just like DOS did before the Mac. But nothing’s more intuitive than face to face.
After two years of research, today we’re launching Atmee v2: a real-time avatar model that makes live video agents feel truly life-like.
• Any character from 1 image + 10s of audio
• Contextual expressions, gestures, even interacting with the environment
• Live video calls, voice only, or video generation
At its core, Atmee v2 is a video foundation model, it can render anything, but it’s built to excel at characters, which is what video models like Seedance 2.5 and Wan 3.0 are mostly used for. In comparison:
Atmee v2
• Up to 50x faster
• 10-100x cheaper
Build tutors, sales coaches, support agents, companions, game characters, even whole living worlds. The interface after text and voice starts today.
Give your agent a document you already use, and it builds new ones in the same layout, down to your logo. It drafts it, renders a sample, and checks it against your example.
Hand it your template, get back your branding.
More ways to interact with your agents on @VeWorldAI.
Explore interactive UI responses, generate documents that match your style, and enjoy a smarter, more intuitive agent experience.
We asked about AI FOMO earlier this week, and one reply stuck with us.
Behind some of those impressive multi-agent setups is a lot of checking outputs, fixing mistakes, managing permissions, and updating things when they break.
The user ended up simplifying things: one well-scoped agent handling one recurring task. A good reminder that you don't have to build an entire AI workforce on day one.
Start with one thing. Experiment, see what works, and build from there.
Plot twist: Chess Coach was built with the help of another AI agent.
We're entering agentception territory. ♟️
Think you can stump Chess Coach?
app.agent.veworld.ai/explore…
Try asking: “Why shouldn't I bring my queen out early if she's my strongest piece?” or throw it something tougher.
A good assistant doesn't wait to be asked. Now your agent doesn't either.
Set a task once, pick a time, and it runs on its own: a market brief at 7am, a weekly summary, a reminder that always lands. Watch it work 👇
A lot of people decide they're "just not a chess person." Usually they just never had a patient teacher.
Chess Coach is one: it explains one move at a time, in plain words, and never sighs. Try it, or build a coach for whatever you'd teach 👇
Anyone else getting a little AI FOMO seeing others run entire teams of agents?
People on X talking about running 20, even 50+ AI agents. Impressive, if it's really working for them. But it makes you wonder: how much of that does the average person actually need?
Not everyone needs a team of agents working around the clock. Sometimes, one agent that takes a recurring task off your plate is enough.
Curious how many agents people actually use regularly, and how much is actually helping.
More ways to interact with your agents on @VeWorldAI.
Explore interactive UI responses, generate documents that match your style, and enjoy a smarter, more intuitive agent experience.
Made some changes to your AI agent and immediately regretted them?
We have version iterations on VeWorldAI.
Make changes, try new things, and roll back when you change your mind.
Because building AI agents should come with a little room for second thoughts.
Meta Muse 🤝 Grok Bot
Two big names stealing the AI agent spotlight lately.
But forget the feature comparisons for a second.
If you're managing a brand or running social media, here's how you might actually put them to work, plus a couple of prompting tricks worth stealing 👇
Grok Bot is your community radar.
Use X convos to spot questions people are asking in your niche.
But instead of chasing the biggest most viral posts, look for genuine questions that haven't been answered well. Cause sometimes the best opportunity isn't where everyone's talking, but WHERE someone could actually use your help.
(That also means resisting the urge to plug your product everywhere 😂)
Two prompts you could try:
Muse:
“Compare my top 3 campaigns by [goal/metric] over the last 30 days with a similar underperformer. Show the differences, flag uncertainties, and suggest ONE experiment.”
Grok Bot:
“Find 5 recent X questions about [topic] that our expertise can help answer. Link the posts, explain why they're worth engaging with, and draft helpful, non-promotional replies.”
And always ask: what evidence supports the recommendation, and what might change it?
