RAG is not broken, naive vector-only RAG fails at scale. What’s broken is the idea that meaning = vector similarity. At scale, that collapses. The future is neurosymbolic AI:
•ontologies for structure
•deterministic layers for truth
•LLMs for interpretation (not
LLMS
-
Beyond Vector Similarity: Neurosymbolic AI and RAG Evolution
By
–
-
Google’s Growing Memory RNNs: New Caching Approach
By
–
"Memory Caching: RNNs with Growing Memory" Google's new paper proposes a simple way to give recurrent models a memory that grows with sequence length. So instead of forcing an RNN to compress the full past into 1 fixed hidden state, it caches memory checkpoints across
-
Malicious Intermediary Attacks on LLM Agent Supply Chain Security
By
–
"Your Agent Is Mine: Measuring Malicious Intermediary Attacks on the LLM Supply Chain" The danger of agent security can also exist in the API router, that is between the agent and the provider. As these routers can read prompts, keys, and tool calls in plaintext, even rewrite
-
Agent Recursion Doubles Usage Through Composable Proactive Systems
By
–
TIL @cognition usage has ~DOUBLED globally since these 2 launches. people are finding all sorts of creative usecases when u can compose agents together and make them proactive. agent recursion is all you need?
-
Gemini 3.1 Flash Lite: Lightning Fast and Super Cheap
By
–
Gemini 3.1 Flash Lite lightning fast and super cheap
-
Kimi 2.6 Code Release Expected Soon
By
–
An image circulating on the web. Looks like Kimi 2.6 code is incoming!
-
LLMs Defer to Official Web Surfaces When Describing Brands
By
–
Amanda: An increasing portion of customers are getting their first exposure to you through a chat with an AI agent you don't control. Me: Generally speaking my experience is leading LLMs are *strikingly* deferential to one's official web surfaces in describing one, by default.
-
Anthropic and OpenAI: Decoding Strategic AI Model Names
By
–
At face value, the claim sounds like a safety warning. It also seems to signal something unusually capable, consequential and maybe beyond what rivals have. It’s hard not to notice the names @AnthropicAI and @OpenAI are using:
— "Spud" (still just a codename) calls to mind -
Hermes Fitness Skill: Exercise Search and Nutrition Calculator
By
–
been using hermes by @NousResearch on telegram and was at the gym today trying to find exercises that target my lower lats — realized there was nothing for fitness on there so i put together a skill that lets hermes search 690+ exercises by muscle group or equipment, look up calories and macros for pretty much any food, and calculate things like BMI, daily calories, one-rep max and body fat % also added a nutrition script so you can compare foods side by side like chicken vs eggs curious to see how it works out when it gets integrated 🏋️, @Teknium github.com/NousResearch/herm… Misturi (@0xmisturi) Finally set up my Hermes Agent today by @NousResearch and i am loving it already — https://nitter.net/0xmisturi/status/2043089306688921785#m
→ View original post on X — @scobleizer, 2026-04-13 18:43 UTC
-
Anthropic Mythos Sets New AI Competition Bar for OpenAI
By
–
Now that @AnthropicAI has introduced Mythos, the bar is set for @OpenAI and “Spud.” In today's @BigTechnology newsletter, @Kantrowitz look at what comes next. We also ask series of questions — like whether “too dangerous” is a new trend that's part warning & part marketing.