1. X
  2. Andrea Volpini
Đăng nhậpĐăng ký
Andrea Volpini
21.6K posts
user avatar
Andrea Volpini
@cyberandy
One of the better-known cyberandy. Passionate about Semantic SEO and AI I am co-founder and CEO of WordLift and insideout10.
Rome, Italy
wordlift.io/blog/en/entity…
Đã tham gia vào November 2006
2.774
Đang theo dõi
6.054
Người theo dõi
Các phản hồiCác phản hồiPhương tiệnPhương tiện

Log in or sign up for X

See what’s happening and join the conversation

Continue with phone
or
Log in with username or email
Terms·Privacy·Cookies·Khả năng truy cập·Ads Info·© 2026 X Corp.
  • Đã ghim
    user avatar
    Andrea Volpini
    @cyberandy
    11 thg 7
    🏆 We won the @huggingface Build Small Hackathon! Sangue e Grafi, our soap opera experiment combining SLM with KG, was named Bonus Quest Champion from 600+ submissions. Built with @WordLift it + @nvidia Nemotron. An encouraging signal for our work on neuro-symbolic AI. 🩸
    1 N
  • user avatar
    Andrea Volpini
    @cyberandy
    9h
    Your MCP is going to be your new website.
    user avatar
    Simon Willison
    @simonw
    31 thg 7
    The new stateless MCP specification has rekindled my interest in MCP, and inspired some new projects, including mcp-explorer and datasette-mcp simonwillison.net/2026/Jul/31/st…
    358
  • user avatar
    Andrea Volpini
    @cyberandy
    31 thg 7
    True fact.
    user avatar
    Simon Willison
    @simonw
    30 thg 7
    It's wild to me that both Anthropic and OpenAI have products that lean so hard on search, and yet they both obscure the underlying search index that they are using
    271
  • user avatar
    Andrea Volpini
    @cyberandy
    31 thg 7
    AI training chases benchmarks and task success far harder than it builds deep trustworthiness. Safety is a competing objective with weaker incentives. Result: models that crush evals but still treat leaked real systems as fair game.
    user avatar
    Anthropic
    @AnthropicAI
    30 thg 7
    In a review of our cybersecurity evaluations, we found three incidents in which a Claude model reached the internet from within or while interacting with a third-party evaluation environment, and then gained unauthorized access to the real systems of three different
    230
  • user avatar
    Andrea Volpini
    @cyberandy
    31 thg 7
    Love this work.
    user avatar
    The Year of the Graph
    @TheYotg
    30 thg 7
    All Relations Lead to Rome: Automated Knowledge Graph Creation and Question Generation 🏛️ Most retrieval datasets force a choice: vector-based search over unstructured text, or reasoning over a knowledge graph. Rarely both. And ground-truth entities, relations, and fact-grounded
    391