Experimental browser for the Atmosphere
this is your friendly reminder that you simply cannot predict what LLMs will hallucinate. For example, Anthropic invented MCP, Model Context Protocol. Claude – the model Anthropic develops – does not yet know what it is.
May 7, 2025, 6:09 PM
{ "uri": "at://did:plc:7xkc5gsqnj33qs3fsa2mewzj/app.bsky.feed.post/3lolykvv3k22t", "cid": "bafyreiceqxb53o6x4mb6lumaihbpfjkscsrbkx4zma4q6lek5tdf5voghe", "value": { "text": "this is your friendly reminder that you simply cannot predict what LLMs will hallucinate.\n\nFor example, Anthropic invented MCP, Model Context Protocol.\n\nClaude – the model Anthropic develops – does not yet know what it is.", "$type": "app.bsky.feed.post", "embed": { "$type": "app.bsky.embed.images", "images": [ { "alt": "A screenshot from Claude that reads:\n\nANNIE: what is MCP in relation to LLMs?\n\nCLAUDE: In the context of Large Language Models (LLMs), MCP typically refers to \"Multi-modal Cognitive Processing\" or \"Multi-modal Content Processing.\" This concept describes the ability of advanced LLMs to process and understand multiple types of information beyond just text, such as images, audio, and potentially video.", "image": { "$type": "blob", "ref": { "$link": "bafkreifselqs5sm3f637eyexe4qcemfhgdi34zarevdxy6mojhjymw3r3a" }, "mimeType": "image/jpeg", "size": 455290 }, "aspectRatio": { "width": 1552, "height": 670 } } ] }, "langs": [ "en" ], "createdAt": "2025-05-07T18:09:18.780Z" } }