Introducing the FLAP LiveMeme Vault 🎬🦋 A new AI-powered Vault that turns static memes into living, evolving stories. Built on Flap's latest AI video infrastructure, LiveMeme lets anyone direct the next scene, with character consistency and story continuity across AI-generated clips. How it works: • Launch a token with a character image and an opening line • Submit the next scene for 0.01 BNB (paid directly to Flap's video service) • AI generates the next clip, building on the previous scene • 100% of trading taxes flow into the Vault, with standard Dev Fees deducted before funding the reward pool • Successfully direct a scene and earn 1%–5% of the reward pool in the token's Quote Asset How rewards work: • Reward rates start at 1% and increase by 1% per interval, up to 5% • Interval duration depends on the scene: 2 min for Scene 1, 3 min for Scene 2, 4 min for Scene 3, and 5 min from Scene 4 onward • Rewards are credited to successful Directors and can be claimed anytime • Token owners cannot withdraw reward pool funds or alter the storyline Each new scene becomes the starting point for the next. Trading fuels creativity. The community keeps the story alive. Explore LiveMeme → 🦋 AI Video Infrastructure docs 👇 Introducing FLAP LiveMeme Vault（活着的模因）🎬🦋 全新 AI 驱动的 Vault 模板，让静态 Meme 变成由社区共同创作、不断演进的动态故事。 基于 Flap 最新的 AI 视频基础设施，任何人都可以成为下一幕的导演，通过 AI 视频生成，让角色与剧情在一次次接力中持续发展。 玩法介绍： • 创建代币：上传一张角色图片，写下一句故事开场白 • 接力剧情：任何人都可以支付 0.01 BNB，提交下一幕剧情（费用直接支付给 Flap 视频服务，不进入奖励池） • AI 生成：基于角色图片和上一段视频的最后一帧生成新片段，让故事持续发展 • 资金积累：100% 的交易税进入 Vault，扣除标准开发者费用后，其余全部进入奖励池 • 导演奖励：成功生成视频的参与者成为新一任 Director，可获得奖励池 1%–5% 的奖励，以代币对应的计价资产发放 奖励机制： • 奖励比例从 1% 开始，每经过一个时间档位增加 1%，最高 5% • 档位时长根据剧情幕数决定：第 1 幕每档 2 分钟，第 2 幕每档 3 分钟，第 3 幕每档 4 分钟，第 4 幕起每档 5 分钟 • 奖励按提交时对应的比例计算，成功生成后记入导演账户，可随时领取 • 视频生成失败不获得奖励，也不会改变当前导演 • 项目方无法提取奖励池资金，也无法擅自修改剧情 每一幕新剧情，都会成为下一位创作者的起点。 交易为创作提供动力，社区让 Meme 的故事持续生长。 立即探索 → 🦋 AI 视频基础设施文档 👇

📢 This Week’s BillionLive Creator Focus: The TapeOut Ecosystem The TapeOut Genesis Transistor Hackathon, hosted by @Ignixbot and X Layer, has officially concluded. Projects explored circuit mining, AI, gaming, music, DeWEB, and more. Meanwhile, @Blonskr has launched TapeCode, supporting the full process from code generation and security audits to DeWEB deployment, making it easier to turn ideas into working applications. This week, we invite everyone in the TapeOut ecosystem to go live on BillionLive. Whether you are a winning team, a hackathon participant, or a creator exploring TapeOut, you can share your experience, demo your project, explain TapeOut, TapeCode, and DeWEB, showcase onchain applications, or discuss your next steps and new opportunities across X Layer. Related livestreams and content will receive additional weight on this week’s Creator Leaderboard. Event page: 📢 本周 BillionLive 创作者激励热点：TapeOut 生态 由 @Ignixbot × X Layer 联合举办的 TapeOut Genesis 创世晶体管黑客松正式收官，参赛项目涵盖电路挖矿、AI、游戏、音乐和 DeWEB 等多个方向。与此同时，@Blonskr 正式推出 TapeCode，支持从代码生成、安全审计到 DeWEB 发布的完整流程，让更多人可以把想法真正做成应用。 本周欢迎所有 TapeOut 生态参与者来到 BillionLive 开播。无论你是获奖团队、参赛开发者，还是关注 TapeOut 的创作者，都可以复盘参赛经历、现场演示项目，分享 TapeOut、TapeCode 与 DeWEB 的基础知识和实操，也可以展示挖矿、AI、游戏、音乐等链上应用，聊聊项目后续规划与 X Layer 生态的新机会。 相关直播与内容将获得本周创作者榜额外权重。 活动链接：

AI-related borrowing is skyrocketing. US Dollar-denominated AI-related corporate bond issuance is up to $419 billion so far in 2026, the highest annual figure on record. This represents a +90% YoY increase from full-year 2025 and a +437% increase from full-year 2024. Year-to-date, Alphabet, $GOOGL, Amazon, $AMZN, Meta, $META, Microsoft, $MSFT, and Oracle, $ORCL, have issued a combined $157 billion in corporate bonds, accounting for 37% of the total. This is followed by investment-grade and high-yield data center and computing companies, which have raised $66 billion and $37 billion, respectively, and together account for ~25% of AI-related issuance. Other AI-linked corporate bonds have represented $159 billion, or 38%. The AI revolution is driving an unprecedented wave of corporate borrowing.

AI agents can reason over enormous amounts of information, but much of the world's actual work is still trapped in PDFs, spreadsheets, scans, and other formats that models struggle to reliably understand. Reducto co-founder and CEO Adit Abraham argues that solving this document layer is essential if agents are going to take on consequential work in healthcare, legal, finance, and insurance. Adit joins a16z General Partner Jennifer Li to explain why traditional OCR breaks down on messy real-world documents, how Reducto combines vision-language models with specialized infrastructure to reach high accuracy without sacrificing cost or latency, and why frontier models can silently drop information on long extraction tasks. Reducto now processes billions of pages each month, and crossed its 7 billionth page the morning of this recording. 0:00 Why AI agents need to read PDFs 2:12 From 1990s OCR to vision models 5:00 Training R1 on the messiest docs 7:46 When agents beat humans at extraction 9:45 A penny a page for 7 billion pages 12:47 Why legal and finance lag coding AI 15:01 The stakes of AI errors in healthcare 16:25 Agents that can edit Word and Excel 17:58 Why the labs won't go deep on PDFs 20:14 Frontier models silently drop data 22:21 Benchmarks have become marketing 25:45 Why charts and Excel stump models 28:46 Infra for 10M documents at once 31:03 How "chet pain" gets claims denied 34:32 The paperwork AI should kill first YouTube: @aditabrm @reductoai @JenniferHli

AI can be civilization-defining and still be overpriced. 🫧 On Bits + Bips, @Nick_van_Eck argues frontier labs like Anthropic have no moat, while @coinbase's John D'Agostino calls AI "a massively productive bubble." With @austincampbell and @perkinscr97 🎧