DEVA-1 is here. Our latest visual synthesis model transforms how digital humans interact, bringing volumetric depth, lifelike expressions, and real-time responsiveness to enterprise workflows. Get access: unith.ai/lp/deva-1 #DEVA1 #UNITH #ConversationalAI #DigitalHumans

Technology is at its best when it helps people connect. 🤝 That's one of the reasons we're excited to become part of @TechBarcelona   . 🚀 As we continue building digital humans designed for real conversations, we're looking forward to joining a community that brings together some of the most innovative companies and minds in Barcelona's tech ecosystem. We're excited to meet new partners, exchange ideas, and keep building technology that feels more human. 💡 Thank you, Tech Barcelona, for welcoming us. Built to speak like a human. Designed to last. #TechBarcelona #DigitalHumans #AI #BarcelonaTech

🚀Meet DEVA-1 by UNITH🚀 Watch the exclusive first teaser right now. This is our next generation of Digital Humans. The most expressive, realistic, and micro-detailed tech we have ever built. We are giving AI a soul by mastering everything a face communicates before the words even start. This is just Drop 01🔥 We have so much more coming that your feed isn't ready for it. #DEVA1 #AIRevolution #DigitalHumans #UNITH #NextGenAI

📊 A 2,000-person company spends the equivalent of 2 full-time employees answering the same 40 questions. That’s not a random estimate. It’s the average we calculate with clients before starting a project. The IT questions. The HR questions. The Legal, Compliance, and Product questions. The ones that “only take 2 minutes” but happen 80 times a day. The cost isn’t just time.⌛ It’s the opportunity cost of the people who know the answers and spend their day repeating them instead of doing the work only they can do. A conversational AI avatar doesn’t solve this because it’s “innovative technology.” It solves it because it’s: 🌍 Available 24/7 💬 Consistent in every answer 🔄 Easy to update when information changes And this is the part most companies underestimate, it records exactly which questions your organisation asks most often. Because the most repeated questions are a direct reflection of what your organisation still doesn’t have clear: ❌ Processes that aren’t documented ❌ Internal communications that aren’t landing ❌ Training that isn’t working That matters to HR. To L&D. To Product teams. To leadership. We’re not selling automation. We’re selling visibility into something that was already happening inside your company — except nobody was measuring it before. ⏳ Have you measured how many hours your team spends answering repetitive questions? #EnterpriseAI #HRTech #DigitalTransformation #ConversationalAI #OperationalExcellence

📊 A 2,000-person company spends the equivalent of 2 full-time employees answering the same 40 questions. That’s not a random estimate. It’s the average we calculate with clients before starting a project. The IT questions. The HR questions. The Legal, Compliance, and Product questions. The ones that “only take 2 minutes” - but happen 80 times a day. The cost isn’t just time. It’s the opportunity cost of the people who know the answers and spend their day repeating them instead of doing the work only they can do. A conversational AI avatar doesn’t solve this because it’s “innovative technology.” It solves it because it’s: 🌍 Available 24/7 💬 Consistent in every answer 🔄 Easy to update when information changes And this is the part most companies underestimate it records exactly which questions your organisation asks most often. That data is valuable on its own. Because the most repeated questions are a direct reflection of what your organisation still doesn’t have clear: ❌ Processes that aren’t documented ❌ Internal communications that aren’t landing ❌ Training that isn’t working That matters to HR. To L&D. To Product teams. To leadership. We’re not selling automation. We’re selling visibility into something that was already happening inside your company, except nobody was measuring it before. ⏳ Have you measured how many hours your team spends answering repetitive questions? #EnterpriseAI #HRTech #DigitalTransformation #ConversationalAI #OperationalExcellence

🌍 More employees = more countries = more languages. That was the challenge: updating the entire sales force when the main product catalogue changed. The previous process looked like this: 📩 Translated materials sent by agencies 🌐 Emails across regions 🏢 In-person training sessions by market ⏰ Manual reminders 📋 Follow-ups managed by local HR teams Average time until 80% of the sales team was fully trained: 11 weeks. With the AI avatar in production: 🌎 The same avatar, available in all 6 operational languages 📱 Accessible from any device, with no installation or extra app required 💬 Able to answer real-time questions about the catalogue, not just replay a script 📊 Full completion logs and FAQ tracking for Product and L&D teams The Learning & Development team stopped managing logistics. Instead, they started receiving a weekly report showing the questions most frequently asked by the sales force. And that became the clearest picture they had ever had of the doubts that appear right before closing a sale. That’s not training automation. That’s a role shift for the team managing training. ⏳ How long does it currently take your company to update sales knowledge when the product changes? #SalesEnablement #LearningDevelopment #EnterpriseAI #ConversationalAI #HRTech

2026 is becoming the year the enterprise AI avatar market gets decided. For companies evaluating solutions right now, this changes the conversation completely.  This category is evolving faster than enterprise buying cycles. What looked differentiated in Q1 suddenly has three competitors by Q3. The platforms that will matter long term are not the ones shipping the most features, but the ones built with structural advantages from day one. Because enterprise is not a feature you add later. It’s an architectural decision. Over the last few months, we’ve spoken with organisations that went through difficult evaluations with other vendors. The pattern is usually the same: they chose the best demo, not the strongest foundation.🔐 And those are two very different decisions. At UNITH, we’ve always believed conversational AI for enterprise needs to be built differently: scalable, secure, measurable, and designed for real business environments from day one. Not adapted later from consumer products. #EnterpriseAI #AIAvatar #DigitalTransformation #AIStrategy #ConversationalAI

💬 The question your CFO is going to ask when you request a renewal for your conversational AI contract: “Does this actually work?” You’d better have the answer before they ask it. Because here’s the uncomfortable reality 👇 Most enterprise AI deployments do work… but nobody can properly measure them. If you can’t attribute: ➡️ which channel generated the interaction ➡️ which campaign drove engagement ➡️ which touchpoint created value …then next year’s budget gets approved blindly. That’s exactly why we just launched Custom Data Tags at UNITH.  🏷️ Every digital human deployment now carries a custom tag across the entire system. Meaning? You can now see exactly: 🌐 which website generated the interaction 📣 which campaign drove activity 🏢 which internal platform is actually being used All directly inside your usage logs. For marketing teams → real attribution. For IT teams → full traceability. For leadership teams → actual visibility into ROI. And for the CFO? It’s the answer to the question that always arrives sooner or later: “So… does this actually work?” This isn’t just another technical feature. It’s the difference between: ✅ a pilot that gets renewed ❌ and a pilot that quietly dies during budget reviews. How are you currently measuring ROI in your AI deployments? #ConversationalAI #EnterpriseAI #AIAvatar #DigitalTransformation #ROI

Yesterday, Arnab Dey and Marcel Alcoverro took the stage at the AI Engineers Barcelona meetup to talk about what it actually takes to build real-time interactive avatars. From speech synthesis and 3D motion generation to Gaussian Splatting and production MLOps, they shared some of the technology powering the humans behind UNITH ✨ But beyond the talk itself, the best part was seeing so many people genuinely curious about where this space is heading. A lot of interesting conversations, new connections, and great energy all around 🚀 Big thanks to everyone who came, asked questions, and stayed to chat afterwards. More soon 👀 Here’s a small recap from the event 👇

Tomorrow part of the UNITH team will be at the AI Engineers Barcelona meetup 👀 🎤 Arnab Dey, PhD and Marcel Alcoverro will be giving a talk on: “Building a Real-Time Interactive Avatar Pipeline” A technical deep dive into what it actually takes to turn an image into a real-time photorealistic human avatar ✨ From: • LLM-driven speech synthesis • Autoregressive 3D motion generation • Gaussian Splatting streamed at 60 FPS • To the MLOps behind running a human face in production Excited to share some of the work happening behind the scenes at UNITH and connect with people building in AI 🚀 If you’re attending, come say hi 👋

Elon Musk recently highlighted a fundamental shift: Digital Humans aren't just a "cool feature". They are the next economic frontier. At UNITH, we aren't just building avatars; we are deploying fully autonomous digital coworkers that bridge the gap between AI efficiency and human connection. What does a UNITH Digital Human bring to your team? ⏰ 24/7 Remote Employees: They don't sleep, don't burn out, and are always ready to represent your brand across every time zone. 💬 Human-Like Conversations at Scale: Forget "if/then" chatbots. Our Digital Humans engage in natural, real-time dialogue - complete with expressions and gestures. 🔗 True Autonomy: From handling high-level customer service to automating internal workflows, they act as active participants in your business, not just static interfaces. As Elon said, once you unlock the Digital Human, you unlock a new level of revenue and scalability. The workforce is evolving. Are you ready to hire your first digital coworker?🚀

We’ve just published something new 👀 A piece where we break down a question we get a lot: Why do businesses actually need a Digital Human? Because enterprise AI is evolving fast, but the way people access and interact with it hasn’t caught up yet. And that gap matters more than it seems. In the article, we explore where Digital Humans fit in, and why they’re becoming a key layer in making AI truly usable at scale. 👉 unith.ai/blog/businesses-nee…  We’d love to hear your thoughts in the comments 💭

The UNITH Blog is now live📚 A new space on our website where we’ll be sharing insights, ideas, and perspectives from our team. Content written by our experts, covering: 🔸What we’re building 🔸 How AI Avatars actually work 🔸 Industry trends 🔸 And what’s coming next If you want to go deeper into what we do (and where this is all heading), this is for you. Grab a coffee and take a look☕  unith.ai/blog

Most companies waste hours on first interviews. Here’s another way ⏳ In this demo, Nia conducts a first-stage job interview with a candidate. She asks structured questions, listens to the answers and keeps the conversation flowing. Just like a human interviewer would, but without the scheduling constraints. An AI Avatar can help hiring teams: 📊 Screen candidates consistently across large applicant volumes ⏱️ Make interviews available 24/7 🤝 Ensure every candidate gets the same structured experience 🎯 Free recruiters to focus on deeper conversations and final decisions For candidates, the experience improves too. Instead of filling out forms or waiting for a reply, they can start with a structured conversation that explains the role and guides them through the first step of the process. Interested in exploring how digital humans could support your hiring process? Let’s connect.