Karl Richard

This reflects a fundamental misunderstanding of pharmacology and virology. Hantaviruses cause severe capillary leak and respiratory failure. There is no credible evidence — or even a plausible mechanism at clinically achievable doses — that ivermectin would meaningfully inhibit hantavirus infection in humans. The absence of evidence that ivermectin doesn’t work is not evidence that it does. That’s not how science works. Promoting unsupported therapies during outbreaks fuels misinformation and can delay care that actually saves lives.

Gilmar Mendes pede a inclusão de Romeu Zema no inquérito das Fake News por causa deste vídeo satírico 👇🏻

@hoje_no @grok quais foram os ataques de Donald Trump ao Papa Leão XIV?

A primeira-ministra da Itália, Giorgia Meloni, reagiu ao terceiro dia de ataques de Donald Trump ao Papa Leão XIV. "As declarações de Trump ao Papa foram inaceitáveis" #hojenomundomilitar

24 hours of mempalace being open source. Here's what happened: The dev community tore it apart. This is how open-source projects can improve: AAAK was labeled "30x lossless compression." It's lossy summarization. Fixed. Token counts used len(text)//3 instead of a real estimator. Fixed. LoCoMo 100% was meaningless - top-k exceeded the corpus size. Acknowledged. The 96.6% LongMemEval is real but it's session-level retrieval on ~50 candidates - not comparable to turn-level numbers in published papers. Point taken, we are not publishing a paper. The benchmark scripts use vanilla ChromaDB. The palace structure only contributed a small amount to that retrieval score of the benchmark. Yet, here is what actually works: verbatim storage, semantic search, MCP integration, fully local, zero API cost. That's the product: All free. All local. All open-source. We also got: 14k stars 2k forks 70+ PRs from contributors 11 bugs fixed and merged so far And... an independent BEAM 100K benchmark showed 49% on answer quality - because mempalace is a retrieval layer, not a reasoning engine! Good news is that we shipped a public correction within hours, not weeks. We're going to keep doing that as much as possible. Thanks to everyone who contributed so far. github.com/milla-jovovich…

Claude Code doesn't show you how many tokens you're using for subscriptions. No breakdown by model. No breakdown by project. Just a progress bar that says "63% used." So I built a local dashboard that reads the files Claude Code already writes to your machine. Turns out every session, every turn, every token is logged to ~/.claude/projects/ in JSONL files. Input tokens, output tokens, cache reads, cache creation, model name, timestamp. It's all there. You just can't see it. My numbers over the last 30 days: 440 sessions. 18,000 turns. $1,588 in API-equivalent costs. On one day, the cache spiked to 700M tokens - visible cache bug, two days in a row. The dashboard scans those local files, builds a SQLite database, and serves charts on localhost:8080. Filter by model (Opus, Sonnet, Haiku). Filter by time range (7d, 30d, 90d, all time). Cost estimates based on current Anthropic API pricing. Works retroactively. First run processes your entire Claude Code history. Install: git clone github.com/phuryn/claude-… cd claude-usage python3 cli.py dashboard Windows: use python instead of python3. Zero dependencies. Python standard library only. Open source, MIT. Star it. Fork it. Make it your own.

@HowToAI_ Very Nice, thanks for sharing! This will come in handy with the LLM Wiki system.

🚨 Someone just open-sourced a tool that converts pdfs to markdown at 100 pages per second. It's called OpenDataLoader. It runs entirely on CPU and handles complex layouts, tables, and nested structures like a senior dev 100% Free.

i can spot a grifter from miles away. so i digged into the code to figure out if this is legit or not. guess i was right. ben is a crypto founder who runs some weird bitcoin lending platform, i was pretty sure he knows absolutely nothing about ai and memory so i tracked down the repo myself since i was curious. his website says he likes to build ai powered products and train local ai models? sure man, 80% of your github repo's are bitcoin related stuff. only one ai related project came up you forked in 2024. mempalace has 10k github stars, more than 1k forks but only.. 7 commits ? apparently the best memory layer to date? no git author history, no account connected to whoever wrote the code of this codebase. it doesn't add up.. the account who pushed the original repo, named: aya-thekeeper, under aya-thekeeper/mempal got deleted right after the repo got published. you paid a random guy named lu to build this shit out for you. ( "Written by Lu (DTL) — March 24, 2026. For: Ben." ) - benchmark md file. lu wrote the code. lu wrote the benchmarks. lu is nowhere in the readme. or mentioned in the github history? the git history then got squashed to one commit and published under milla jovovich? seriously? a actress? you say she is a great friend of yours, she has been building this project with you. she does this at night. yet she has.. 7 commits and only 2 active days in her entire github history? you paid an actress and a random guy to promote a product you know absolutely nothing about.

When Andres Freund, Linux kernel contributor & Microsoft engineer was debugging slow SSH logins on his Debian machine in March 2024, he noticed something weird: liblzma (part of XZ Utils) was using way too much CPU power, so he kept digging, and what he uncovered was a multi-year supply-chain attack! An attacker using the name “Jia Tan” had spent two years slowly infiltrating the tiny XZ Utils project, a compression library used by virtually every major Linux distribution. The backdoor wasn’t in the source code. It was hidden deep inside the build scripts. It would have given the attacker remote root access on millions of servers the moment a specially crafted SSH key was used. Freund caught it days before it would have shipped in Debian, Fedora, Ubuntu and more. One man, one anomaly, one routine debug session saved the internet from a potential catastrophe. Respect!

My friend Milla Jovovich and I spent months creating an AI memory system with Claude. It just posted a perfect score on the standard benchmark - beating every product in the space, free or paid. It's called MemPalace, and it works nothing like anything else out there. Instead of sending your data to a background agent in the cloud, it mines your conversations locally and organizes them into a palace - a structured architecture with wings, halls, and rooms that mirrors how human memory actually works. Here is what that gets you: → Your AI knows who you are before you type a single word - family, projects, preferences, loaded in ~120 tokens → Palace architecture organizes memories by domain and type - not a flat list of facts, a navigable structure → Semantic search across months of conversations finds the answer in position 1 or 2 → AAAK compression fits your entire life context into 120 tokens - 30x lossless compression any LLM reads natively → Contradiction detection catches wrong names, wrong pronouns, wrong ages before you ever see them The benchmarks: 100% recall on LongMemEval — first perfect score ever recorded. 500/500 questions. Every question type at 100%. 92.9% on ConvoMem — more than 2x Mem0's score. 100% on LoCoMo — every multi-hop reasoning category, including temporal inference which stumps most systems. No API key. No cloud. No subscription. One dependency. Runs on your machine. Your memories never leave. MIT License. 100% Open Source. github.com/milla-jovovich…

@hoje_no Tá parecendo as ruas aqui de São Paulo, com as crateras fazendo aniversário.

O resultado dos ataques aéreos à base iraniana de Hamadan, uma das mais importantes do regime. Com esses danos, a base ficou temporariamente inoperacional.

US intelligence assesses that Iran's supreme leader, Ayatollah Khamenei, is dead -FOX News

Brazil just raised the tariffs of GPUs, CPUs, and RAM from 0-2% to 10-20%. The country produces zero (0) of these, but has abundant clean electricity to host datacenters. Couldn't think of a more self-defeating policy at this time.

@hoje_no Ele está se apoiando num andador?

Eric Slover, que recebeu ontem a Medal of Honor por suas ações na captura do ditador Nicolás Maduro, é a personificação do soldado profissional autêntico. Só pelo fato de ter a patente Chief Warrant Officer 5 (CWO 5) já é algo muito raro. No ombro direito, ele tem o distintivo de combate do Comando de Operações Especiais, a "Delta Force" da aviação. Foi o primeiro CWO 5 na história a receber a Medal of Honor e tem 10 distintivos de combate, o que significa que ele esteve, no mínimo, 5 anos em ambiente de combate. Em resumo, o cara é brabo!

O ministro Alexandre de Moraes mandou INTIMAR o presidente da Unafisco por críticas à operação da PF que mirou servidores da Receita Federal. cnnbrasil.com.br/blogs/caio-jun…

Quando autoridades não conseguem desmentir dados de contratos suspeitos ou movimentações familiares de dinheiro, elas partem para a acusação de que eles foram acessados ilegalmente. É o método Flávio: os dados são reais e indicam a sujeira, mas… não podiam ter sido descobertos assim. Anule-se tudo e fim de papo. Finja-se que nada aconteceu. O Brasil segue preso num looping de 2019.

Em 2023, Toffoli anulou, sozinho, todas as provas do acordo de leniência da Odebrecht. Ele incluiu na fundamentação de sua decisão um parecer do Departamento de Recuperação de Ativos e Cooperação Internacional, à época subordinado a Augusto de Arruda Botelho, então secretário nacional de justiça. O parecer dizia que não havia registro do pedido de cooperação do MPF à Suíça, requisito legal para uso das provas. ANPR e PGR, contudo, comprovaram, em seus recursos, que havia sim o pedido formal. O DRCI admitiu a existência e finalmente encontrou o registro. Apesar disso, os recursos jamais foram sequer analisados, permitindo que centenas de corruptos, inclusive confessos, fossem beneficiados no Brasil e em vários outros países. Botelho havia sido advogado da Odebrecht e, recentemente, já como advogado do diretor de compliance do Master (alvo da Operação Compliance Zero), viajou em jato particular com o min. Toffoli, relator do caso. Em outubro de 2023, a Transparência Internacional (TI) denunciou a decisão infundada de Toffoli ao Grupo de Trabalho da OCDE (WGB/OECD). O órgão inseriu o caso no topo da lista de verificações sobre o Brasil, na revisão que estava realizando sobre o cumprimento de sua Convenção contra o Suborno Transnacional. Em março, o Brasil terá que apresentar explicações e atualizações à OCDE sobre a anulação das provas da Odebrecht, que configura violação gravíssima às obrigações assumidas pelo país junto à convenção da OCDE. Apenas três meses depois, publicamos, junto com o Índice de Percepção da Corrupção (IPC) de 2023, um relatório que destacava criticamente a atuação de Toffoli no caso da Odebrecht. Dois dias depois, o ministro determinou abertura de investigação contra a TI. Apesar da PGR ter promovido o arquivamento do inquérito por total falta de provas e de competência de Toffoli, o ministro ignorou a posição da PGR e mantém aberto o inquérito até hoje. O caso já foi citado duas vezes pela Comissão Interamericana de Direitos Humanos (CIDH) da OEA como tentativa de criminalização da Transparência Internacional. Agora, junto ao IPC 2025 recém-lançado, reiteramos nossas denúncias e recomendamos que Toffoli seja afastado e investigado. O Brasil e a comunidade internacional precisam acompanhar com máxima atenção este caso. A corrupção do Judiciário impede o desenvolvimento socialmente justo e é devastadora para os direitos humanos, a segurança pública e a democracia. A luta contra a corrupção não é um fim em si mesma, é uma luta por direitos.

A Polícia Federal aponta aquilo que o Brasil inteiro sabia: a suspeição de Toffoli no caso Master. Ou os senadores, os demais membros do STF e a PGR retiram o ministro imediatamente do cargo ou passam também a ser suspeitos.

@MaziyarPanahi @huggingface @OpenMed_AI @grok o que é OpenMed?

🚨 OpenMed just expanded the PII detection arsenal. 105 language-specific models for French, German, and Italian, now open source. All Apache 2.0. All free. Forever. European healthcare AI just got a lot more accessible. Supporting GDPR, HIPAA, and privacy compliance: 🇫🇷🇩🇪🇮🇹

@hoje_no Hahaha que ótimo.

Muito raramente, de tempos em tempos, aparece algum russo na TV estatal do Kremlin falando a realidade sobre o que está efetivamente acontecendo no campo de batalha. Neste vídeo vemos um falando que a Ucrânia não pode ser derrotada e que a Rússia está passando por grandes dificuldades. Podemos notar o desespero dos outros convidados em tentar cortá-lo.

Acabei de perguntar ao presidente Hugo Motta se o projeto que estava sendo votado faria com que funcionários da Câmara passassem a ganhar acima de deputados, extrapolando o TETO e ele respondeu que sim 👍 Esse é o tuíte.