Live from Code with Claude London: we're launching self-hosted sandboxes (public beta) and MCP tunnels (research preview) in Claude Managed Agents.
Run agents inside your own perimeter, with your security controls applied by default.
@NewsNation@timburchett@timburchett ... The arrogance comes from the fact that unelected bureaucrats can "keep the gates" far longer than elected politicians. It is sad, but a fact that is seen too many times.
@DrBeaVillarroel I’d like to know if anyone has taken another look at the spots where those old transients showed up, just to see if anything similar is still happening there. Is anyone already checking this systematically, maybe the Galileo Project (Avi Loeb)?
@DrBeaVillarroel You can already observe this behaviour in children. If I can't have it, no one else should either. If I can't control it, I'll destroy it.
@skdh I had another exciting experience. People find a small mistake that an AI has made and then no longer trust it at all and even say that it is therefore not possible to work with it in a meaningful way.
@DrBeaVillarroel your work is done and they can’t do anything better than this - it is perfect for marketing of your amazing scientific efforts on this immense importance for humanity at all. what ever you found, it has the potential to change mankind forever.
@HansStarlife@DrBeaVillarroel@PhdBrandenburg I am listening very careful to every interview with Avi Loeb and I hear only his voice for beeing open to all possibilities in science w/o any bias. The “Loeb-Scale” is something that makes sense as well, to give this neutral perspective a serious chance.Evidence and data-driven.
@MChroniken Hallo Sirwan, ich verfolge deinen YT-Kanal schon einige Zeit. Du wirst immer besser und ich freue mich sehr in Österreich einen Weltraum-Enthusiasten zu haben. Mach bitte weiter, werde noch besser und bleib deiner Leidenschaft weiterhin treu. Danke für deine Arbeit!
@FlorianGallwitz … ich habe mich gerade gefragt, welche Haluzinationsrate eigentlich ein Mensch hat? Bei KI wird ja beim kleinsten Ansatz davon sofort die grundsätzliche Vertrauensfrage gestellt. Beim Menschen ist fehlerhafte Information nach meiner Erfahrung aber eher „normal“.