04-07 01:22 · 财经,财经
【Omdia：受社交视频广告推动 2030年全球在线视频和电视收入将超过1万亿美元】Omdia媒体和娱乐部门负责人Maria Rua Aguete在马德里FED展会上发布的最新数据显示，预计到2030年，全球传统电视和在线视频收入将超过1万亿美元。数据凸显了媒体和娱乐行业的重大结构性转变，总收入预计将从2025年的7750亿美元增至2030年的1.03万亿美元，增长主要由数字形式的业务驱动，尤其是广告业务。
Omdia：受社交视频广告推动 2030年全球在线视频和电视收入将超过1万亿美元
Omdia媒体和娱乐部门负责人Maria Rua Aguete在马德里FED展会上发布的最新数据显示，预计到2030年，全球传统电视和在线视频收入将超过1万亿美元。数据凸显了媒体和娱乐行业的重大结构性转变，总收入预计将从2025年的7750亿美元增至2030年的1.03万亿美元，增长主要由数字形式的业务驱动，尤其是广告业务。
（文章来源：界面新闻）
Omdia媒体和娱乐部门负责人Maria Rua Aguete在马德里FED展会上发布的最新数据显示，预计到2030年，全球传统电视和在线视频收入将超过1万亿美元。数据凸显了媒体和娱乐行业的重大结构性转变，总收入预计将从2025年的7750亿美元增至2030年的1.03万亿美元，增长主要由数字形式的业务驱动，尤其是广告业务。
（文章来源：界面新闻）
▸ 展开全文
04-07 01:22 · 财经,财经
【海外与AI相关业务逐渐进入收获期 这家公司已为客户提供数字人民币2.0相关产品】近日，央行新增12家数字人民币业务运营机构，数字人民币业务运营机构总数从原有的10家扩容至22家，数字人民币生态建设迈入新阶段。宇信科技(300674)日前在券商策略会、电话会议中表示，公司已直接为客户提供数字人民币2.0相关产品，项目已进入实质性推进阶段。
近日，央行新增12家数字人数字人民币业务运营机构总数从原有的10家扩容至22家，数字人民币生态建设迈入新阶段。
公司表示，相较于1.0时代聚焦支付接口与钱包开立，2.0阶段需触及银行
宇信科技在跨境支付银行提供相关服务。在跨境支付生态布局方面，公司已与合作伙伴共同探索数字货币
AI业务收入方面，2025年公司实现AI相关收入2亿元，包括AI硬件、AI软件及AIAgent相关业务。公司推出了智算租赁银行提供“智算算力+场景服务”。公司表示，正打造金融AI的超级入口，基于这一安全、合规的云底座，部署了一系列开箱即用的金融AIAgent，使公司从传统的项目制软件开发租赁”的服务模式积极转型。
机构指出，宇信科技业务结构优化，海外商业化落地加速。看好公司的生态卡位及数字人民币2.0下带来的增量机遇。
（文章来源：人民财讯）
▸ 展开全文
04-07 01:22 · 财经,财经
【中使馆提醒：在巴新中国公民和机构切勿从事非法采掘和走私黄金活动】据中国驻巴布亚新几内亚大使馆微信公众号消息，近期，巴布亚新几内亚再次发生中国公民因涉非法采掘和走私黄金被当地执法部门查获、羁押和遣返等事件。
据中国驻巴布亚新几内亚大使馆微信公众号消息，近期，巴布亚新几内亚再次发生中国公民因涉非法采掘和走私黄金
根据巴布亚新几内亚矿产开发相关法律规定，在巴新从事采金须事先取得相应开采许可文件，超许可范围进行开采属违法行为，不受法律保护。非法砂金开采和黄金走私属于严重违法行为。
中国驻巴布亚新几内亚大使馆再次郑重提醒在巴新中国公民和机构切勿从事非法采掘和走私黄金活动，切实增强安全风险意识、确保人身财产安全。从事采金的中国企业应及时向使馆报备，如遇紧急情况第一时间报警并与中国驻巴新大使馆联系寻求帮助。
（文章来源：中国新闻网）
▸ 展开全文
04-07 01:22 · 财经,财经
【鸿海3月营收8037亿新台币创新高 同比增长45.6%】鸿海2026年3月营收为8037亿新台币，环比增长34.90%，同比增长45.57%，为历年同期最高。2026年第一季营收为2.1296万亿新台币，环比下降18.18%，同比增长29.68%，为历年同期最高。
鸿海3月营收8037亿新台币创新高 同比增长45.6%
鸿海2026年3月营收为8037亿新台币，环比增长34.90%，同比增长45.57%，为历年同期最高。2026年第一季营收为2.1296万亿新台币，环比下降18.18%，同比增长29.68%，为历年同期最高。
（文章来源：科创板日报）
鸿海2026年3月营收为8037亿新台币，环比增长34.90%，同比增长45.57%，为历年同期最高。2026年第一季营收为2.1296万亿新台币，环比下降18.18%，同比增长29.68%，为历年同期最高。
（文章来源：科创板日报）
▸ 展开全文
04-07 01:22 · 开源,健康科技,数据分析,隐私保护,本地化AI
Free, open-source blood work dashboard — AI-powered PDF import, DNA raw data analysis (42 SNPs), trend charts, and personalized health insights for 287+ biomarkers. Local-first, no account required.
getbased is a free, open-source health dashboard that turns lab PDFs into interactive charts and AI-powered health insights. Track 287+ biomarkers over time, detect trends, and get personalized interpretations — all stored locally in your browser with no account required.
Live app · Documentation · Discord · Nostr
- AI-powered PDF import — drop any lab report (any format, language, or country) and AI extracts and maps results to 287+ known biomarkers automatically. Batch import, direct image import (JPG/PNG/WebP), auto image mode for scanned PDFs
- Biomarker trend charts — interactive line charts with proportional time scale, reference bands, optimal ranges, and trend detection across 16 categories
- AI chat — ask questions about your results with full health context, image attachments, multiple personalities, conversation threads
- DNA import — upload raw data from AncestryDNA, 23andMe, MyHeritage, FTDNA, or Living DNA. 42 curated SNPs across 10 categories (methylation, iron, lipids, vitamin D, etc.) with APOE haplotype resolution. Genetic factors shown on dashboard, detail modals, and in AI context
- Specialty lab adapters — OAT (165 markers), fatty acids (Spadia, ZinZino, OmegaQuant), Metabolomix+. Any other specialty test imports through the custom marker pipeline
- Biological age — PhenoAge (Levine 2018) + Bortz Age (Bortz 2023) combined into a unified Biological Age marker with component breakdown
- Calculated markers — HOMA-IR, BUN/Creatinine ratio, free water deficit, lipid ratios (TG/HDL, LDL/HDL, ApoB/ApoA-I), NLR, PLR, De Ritis ratio, hs-CRP/HDL cardiovascular risk ratio
- Trend alerts — sudden changes and linear regression flagged on the dashboard
- Correlation viewer — compare any two markers, heatmap view
- Compare dates — side-by-side comparison of any two lab dates
- Manual entry — add results without a PDF, create custom biomarkers
- Marker glossary — searchable reference for all markers with values and ranges
- Interpretive lens — frame AI analysis through specific scientific paradigms or experts
- 9 lifestyle context cards — diet & digestion, sleep, exercise, stress, light & circadian, environment, EMF assessment (Baubiologie SBM-2015), and more — each gets an AI health rating and enriches all interpretations
- Menstrual cycle tracking — phase-aware reference ranges, cycle phase bands on charts, perimenopause detection, symptom tracking
- Supplement & medication timeline — overlaid on charts to correlate with biomarker changes
- PDF reports — export a full health report as PDF
- Multi-profile — track multiple people, client list with search/sort/filter
- All data stored locally in your browser (localStorage + IndexedDB) — nothing on a server
- Personal info stripped from PDFs before AI processing (regex + streaming local AI obfuscation)
- AES-256-GCM encryption at rest
- Automatic backups (IndexedDB snapshots + daily folder backup via File System Access API)
- Venice AI end-to-end encryption option — prompts encrypted client-side (ECDH secp256k1 + AES-256-GCM), decrypted only inside a TEE. Nothing readable in transit or at rest on their servers
- Run a local AI server and nothing leaves your machine at all
- No account, no sign-up, no tracking
Opt-in feature that lets AI agents query your lab context — coding agents (Claude Code, Cursor), messenger bots (Hermes Agent, OpenClaw), or any MCP-compatible tool.
- Enable in Settings → Data → Agent Access to generate a read-only token
- Context is pushed to a lightweight gateway on every save and profile switch
- Per-profile: each profile's context is stored separately; agents can query any profile by ID
- Use getbased-mcp to connect Hermes Agent, OpenClaw, or any MCP-compatible agent
- Only the AI-readable context text is shared — never your mnemonic or raw lab data
- Token is revocable at any time from the same settings panel
Switch providers anytime. All non-AI features work without a provider configured.
git clone https://github.com/elkimek/get-based
cd get-based
node dev-server.js
Open http://localhost:8000
. You need an AI provider API key or local AI server for PDF import and chat. All other features work without one.
No build tools, no bundler, no package manager. Pure ES modules — 36 files under js/
.
- Chart.js for interactive charts
- pdf.js for PDF text extraction
- All dependencies vendored locally (
vendor/
) — no CDN calls - Installable as a PWA (works offline for non-AI features)
29 browser-based test files run headlessly:
./run-tests.sh
Starts a local server, runs all tests via Puppeteer, exits 0/1.
See CONTRIBUTING.md. Project board: planned features.
GPL-3.0. See LICENSE.
▸ 展开全文
04-07 01:22 · 开源工具,AI营销,自动化流程,内容生成,多模型集成
50 AI agent skills for affiliate marketing. Research trending content, write data-backed posts, generate infographics, build landing pages, deploy — full flywheel with social intelligence. Works with Claude Code, Pi, ChatGPT, Gemini, Cursor, Windsurf, any AI.
Turn any AI into your affiliate marketing team.
50 AI-powered skills across 8 stages with a closed-loop flywheel. Research programs, scout trending content, write data-backed posts, generate infographics, build pages, deploy, track, optimize, scale — with any AI agent.
Works with: Claude Code · Pi · ChatGPT · Gemini CLI · Cursor · Windsurf · OpenClaw · any AI that reads text
# Claude Code / Pi (recommended)
npx skills add Affitor/affiliate-skills
# Or clone manually
git clone https://github.com/Affitor/affiliate-skills.git ~/.claude/skills/affiliate-skills
cd ~/.claude/skills/affiliate-skills && ./setup
# OpenClaw / ClawHub
clawhub install affiliate-skills
# Cursor / Windsurf
npx skills add Affitor/affiliate-skills
Paste this into any AI:
Search the Affitor affiliate directory for AI video tools.
Use this API: GET https://list.affitor.com/api/v1/programs?q=AI+video&sort=top&limit=5
Show me the results in a table with: Name, Commission, Cookie Duration, Stars.
Then recommend the best one and explain why.
- You Google "best affiliate programs" and get SEO spam written to rank, not to help
- You write content from gut feeling with no idea what format actually performs
- You have no data on what's trending, what hooks work, or what gaps exist
- You spend 4 hours on a landing page that converts at 0.2%
- You pick programs by vibes instead of data
S1 RESEARCH ──▶ S2 CONTENT ──▶ S3 BLOG & SEO ──▶ S4 OFFERS & LANDING
▲ │
│ ▼
│ S5 DISTRIBUTION
│ │
└──────────── S6 ANALYTICS ◀─────────────────────────┘
│
▼
S7 AUTOMATION → SCALE
│
S8 META (across all)
Closed loop. S6 Analytics feeds BACK to S1 Research. Every skill knows what comes next (chain_metadata.suggested_next
) and what feeds it. Data flows forward through the funnel and back through analytics.
Data-driven. S1 skills don't just find programs — they scout what content is actually performing (views, likes, shares), analyze traffic, and rank angles by real engagement data. By the time you write, you already know what format, hook, and platform to use.
Research-backed. Content skills don't write from thin air. content-research-brief
collects real source articles first. Your posts have real stats, real quotes, and current information.
You: I want to promote AI video tools. What's trending?
AI: [runs trending-content-scout]
Scanned 47 videos across YouTube + TikTok.
Top format: comparison (45% of top content, avg engagement: 35.2)
Best hook: "I replaced my $5K video team with AI" (engagement: 42.3)
Gap: nobody comparing HeyGen vs Synthesia on TikTok
Benchmark: median 18K views, top 10% needs 85K+
You: What's the best program to promote?
AI: [runs affiliate-program-search + traffic-analyzer]
Name Commission Cookie Stars Traffic Score
──────────────────────────────────────────────────────────────
HeyGen 30% recurring 60 days ★ 127 82/100
Synthesia 25% one-time 30 days ★ 84 76/100
D-ID 20% recurring 45 days ★ 52 61/100
HeyGen wins: highest commission, recurring, longest cookie,
strongest website traffic (2.1M monthly visits, 35% direct).
You: Research the latest about HeyGen for a LinkedIn post.
AI: [runs content-research-brief]
📚 7 sources collected (5 fully fetched)
🏷️ Tags: AI (7), Funding (2), Tools (5)
Key stats:
- "$60M Series A at $500M valuation" — TechCrunch
- "40,000+ businesses use HeyGen" — company blog
- "Avatar 3.0 launched with real-time lip sync" — The Verge
3 angles generated:
1. "HeyGen just raised $60M — here's what it means for AI video"
2. "I tested HeyGen vs Synthesia for 30 days — honest comparison"
3. "AI video tools are killing the $45B production industry"
You: Write angle 2 as a LinkedIn post with an infographic.
AI: [runs viral-post-writer + infographic-generator]
Post: 1,200-char comparison post, bold claim hook,
soft CTA in comments, FTC disclosure.
Infographic: side-by-side comparison card (1080×1350),
HeyGen vs Synthesia — commission, features, pricing.
Dark modern palette, self-contained HTML ready to render.
You: Now a TikTok script.
AI: [runs tiktok-script-writer]
45-second script, demo_first hook (backed by scout data):
"Watch me create a $2,000 video for free in 30 seconds."
Hook → Pain → Demo → Result → CTA. Bio link reminder.
You: Plan my full funnel — I have 5 hours/week.
AI: [runs funnel-planner]
Week 1: Scout + Research → find winning angles + collect sources
Week 2: Content → 5 LinkedIn posts + 2 TikToks + 2 infographics
Week 3: Blog → comparison article backed by research brief
Week 4: Deploy → landing page + bio link live
Week 5: Analytics → compare your metrics vs scout benchmark
Week 6: Optimize → double down on what beat the benchmark
Target: first affiliate click by week 2, first commission by week 5.
Affiliates — Find programs, create content, track performance. Data-driven, not vibes.
Influencers & Creators — Research what's trending, create platform-native content, add infographics. Know what format and hook to use before you create.
Marketers — Content research, competitive analysis, multi-platform strategy. Research
▸ 展开全文
04-07 01:21 · 开源竞品,AI助手,隐私安全,本地部署,免费模式
Natively - Free open-source AI interview copilot & meeting assistant. The best Cluely alternative, Final Round AI alternative, and Interview Coder alternative. Real-time transcription, undetectable stealth mode, local RAG, BYOK. No subscriptions. No data breaches.
The best free alternative to Cluely, Final Round AI, LockedIn AI, and Interview Coder.
Same UI as Cluely. More features. $0. Open source. No data breaches.
Competitors charge $20–$149/month, store your data on their servers, and one already breached 83,000 users. Natively costs $0, runs locally, and has never had a data breach. Your keys, your models, your machine.
Requires macOS 12+ (Apple Silicon & Intel) or Windows 10/11
🔥 49.4k views · 💸 $0 vs $149/mo rivals · ⚡ <500ms latency · 🛡️ 0 data breaches
Natively started as a pixel-perfect recreation of Cluely's interface — then kept going. If you've used Cluely, you already know how to use Natively. Same overlay, same workflow, same shortcuts. Except it's free, open-source, runs locally, supports any LLM, and has never breached a single user's data.
Looking for a free Cluely alternative? A Cluely open-source clone? You found it.
Natively works as a free, undetectable AI coding interview assistant for standard online assessments. It captures your screen, analyzes the problem, and gives you real-time hints, solutions, and explanations — all through an invisible overlay that doesn't interfere with your coding environment.
Works undetected on:
- LeetCode (including LeetCode contests)
- HackerRank
- CoderPad
- Codility
- HackerEarth
- Karat
- Any browser-based coding environment
How it works:
- Screenshot the problem with a single shortcut
- Natively OCRs the question and sends it to your chosen AI (GPT, Claude, Gemini, or local Ollama)
- Response appears in the invisible overlay — never on screen share
⚠️ Important: Natively is not designed to bypass dedicated proctoring software like Pearson VUE, ProctorU, or Respondus Lockdown Browser — these run at the OS level and are a different category entirely. For standard online coding assessments without dedicated proctoring software, Natively's stealth mode is not detectable.
- Cluely had a data breach in mid-2025 that exposed 83,000 users' personal info, transcripts, and screenshots — Natively stores everything locally with limited basic telemetry and has never had a breach.
- Final Round AI costs $149/month and its taskbar icon is visible to proctoring software — Natively is free, open-source, and has a battle-tested undetectable stealth mode.
- LockedIn AI charges $55–70/month and locks you into their cloud LLM with no local option — Natively lets you use any model (GPT, Claude, Gemini, Llama) or go fully offline with Ollama.
Every star pushes Natively higher in GitHub search, helping developers and job seekers find a free, private alternative instead of paying $149/month for tools that store their data on someone else's server.
This demo shows a complete live meeting scenario:
- Real-time transcription as the meeting happens
- Rolling context awareness across multiple speakers
- Screenshot analysis of shared slides
- Instant generation of what to say next
- Follow-up questions and concise responses
- All happening live, without recording or post-processing
Legend: ✅ Full support ·
⚠️ Partial or limited · ❌ Not available
The UI is intentionally familiar — if you've used Cluely, there's zero learning curve.
Cluely's mid-2025 data breach exposed personal information, full interview transcripts, and screenshots of 83,000 users. Every word spoken during an interview was stored on their servers — and then leaked. They charge $20/month for this privilege.
Natively has no backend, no servers, and limited telemetry (basic GA4 install tracking, zero user data). Your transcripts, API keys, and screenshots never leave your machine. The entire codebase is open-source (AGPL-3.0) and auditable. Zero breaches, zero data collection — that is the only acceptable standard for a tool that listens to your interviews.
LockedIn AI is the most expensive tool in the category at $55–70/month. It locks you into a single cloud LLM with no option for local inference. Every transcript and response passes through their servers.
Natively supports every major model (Gemini, GPT, Claude, Groq) via bring-your-own-key, and offers 100% offline mode through Ollama. You pay only for the API tokens you actually use — or pay nothing at all by running Llama 3 locally. No subscription, no vendor lock-in.
Final Round AI is the most expensive option at $149/month, optimized for pre-interview prep and mock interviews but with the slowest live latency in the category. Critically, its taskbar icon is visible to proctoring software, making it detectable during monitored interviews.
Natively delivers <500ms end-to-end latency using Rust-based native audio capture with Zero-Copy ABI Transfers. Its undetectable stealth mode hides from the dock, disguises process names, and syncs state across all windows — battle-tested and hardened across five major releases.
Pluely is a solid lightweight alternative (~10MB, Tauri-based) and it has Linux support, which Natively does not yet offer. Credit where it is due.
But Pluely is a basic overlay. It has no local RAG, no meetin
▸ 展开全文
04-07 01:21 · 地缘政治,中东局势,军事冲突,市场波动,供应链风险
Defense Secretary Pete Hegseth says US forces are carrying out more strikes on Iran than any day since start of war.
US President Donald Trump has warned of the “complete demolition” of Iran’s power plants and bridges in a matter of hours if the Strait of Hormuz is not fully reopened by Tuesday’s deadline.
Trump says Iran’s response to the US ceasefire proposal, conveyed through intermediaries, is “significant”, but “not good enough”.
▸ 展开全文
04-07 01:21 · 融资,AI估值,商业模式,市场竞争,技术叙事
The back story behind the first "$1.8B" dollar "AI Company"
The back story behind the first “$1.8 Billion” dollar “AI Company”
AI isn’t the only thing behind Medvi
On Thursday, The New York Times published a thing
and it went viral, declared as a victory for AI:
But there’s a lot more to be said that was only scarcely touched on in the article.
Like this
And this
And this brutal, compelling dissection on YouTube
which it itself built on and extending an earlier (May 2025) dissection from Futurism that The New York Times should have read and addressed in far more detail:
A friend of mine who has been tracking this for a while had sees Medvi as “a fraud-layer on top of also-scammy-but-possibly-less-illegal platforms”, speculating that “If there is any money there, they will be sued by all their suppliers and vendors, because I’m sure they’re in violation of every agreement in terms of compliance efforts, safe data handling, etc.” (The friend also is doubtful of the revenue reports, asking “why would this be the only thing they’re telling the truth about?”)
All in all, glorifying Medvi was not The New York Times’ finest hour.
And hardly the poster child AI boosters should be hoping for. Instead, as the YouTube video author, Voidzilla, notes, if anything Medvi is a warning sign — for how AI can be abused.
▸ 展开全文
04-07 01:21 · 大模型合作,芯片优化,算力竞争,AI基础设施,技术联盟
Anthropic expands partnership with Google and Broadcom for next-gen compute
Anthropic expands partnership with Google and Broadcom for multiple gigawatts of next-generation compute
We have signed a new agreement with Google and Broadcom for multiple gigawatts of next-generation TPU capacity that we expect to come online starting in 2027. This significant expansion of our compute infrastructure will power our frontier Claude models and help us serve extraordinary demand from customers worldwide.
“This groundbreaking partnership with Google and Broadcom is a continuation of our disciplined approach to scaling infrastructure: we are building the capacity necessary to serve the exponential growth we have seen in our customer base while also enabling Claude to define the frontier of AI development,” said Krishna Rao, CFO of Anthropic. “We are making our most significant compute commitment to date to keep pace with our unprecedented growth.”
Demand from Claude customers has accelerated in 2026. Our run-rate revenue has now surpassed $30 billion—up from approximately $9 billion at the end of 2025. When we announced our Series G fundraising in February, we shared that over 500 business customers were each spending over $1 million on an annualized basis. Today that number exceeds 1,000, doubling in less than two months.
The vast majority of the new compute will be sited in the United States, making this partnership a major expansion of our November 2025 commitment to invest $50 billion in strengthening American computing infrastructure.
The partnership deepens our existing work with Google Cloud—building on the increased TPU capacity we announced last October—as well as our relationship with Broadcom.
We train and run Claude on a range of AI hardware—AWS Trainium, Google TPUs, and NVIDIA GPUs—which means we can match workloads to the chips best suited for them. This diversity of platforms translates to better performance and greater resilience for customers who depend on Claude for critical work. Amazon remains our primary cloud provider and training partner, and we continue to work closely with AWS on Project Rainier. Claude remains the only frontier AI model available to customers on all three of the world's largest cloud platforms: Amazon Web Services (Bedrock), Google Cloud (Vertex AI), and Microsoft Azure (Foundry).
▸ 展开全文