Prioritize Misuse Risks Over Superintelligence; Hold Models Back
I think this hangs on whether you see AI risks more from bad actors mis-using AI vs an AI becoming a super-intelligent adversary. Since I'm 99% in the former camp, I see responsible AI companies holding models back until mitigation is in place for potential harms as responsible.
ManiacVC Built as Fully AI‑Native Investment Firm
this is why i chose to build @maniacvc to be AI native from the ground up. all of maniac is AI native from outbound sourcing, analyst work, inbound handling, investment memos, and everything in between.
Ex-Microsoft AI Leader Eric Boyd Heads Anthropic Infrastructure
Eric Boyd, former president of AI platform at Microsoft, has joined Anthropic as head of infrastructure. Boyd had spent nearly 17 years at Microsoft, after originally joining in 2009 and running the Bing Ads dev team https://t.co/SQVDoJmNql
Hinton Unveils Forward-Forward Algorithm in Nobel Lecture
#TechTuesday. December 8, 2024. Geoffrey Hinton delivered his Nobel Prize lecture on #AI and the future titled 'The Forward Forward Algorithm'. #JVGpost https://t.co/R5CtrgkpJ0
LLM Story Feels Polished yet Lacks Logical Cohesion
I think the story that was shared in the Mythos System Card still has the signs of flawed LLM writing (which looks like good writing at first glance): A story that doesn't really hold together logically, but sounds like it...
Future Robots and Humans Will Share a Dynamic Holodeck
Interesting post by @peteflorence who is a robotics pioneer. He triggered me by positioning against people who talk about technology like world models. But he is right. Where we are going with robotics is the interesting part. It is clear to me that both...
AI Adoption Gap: 60% Miss Massive Opportunity
Only 40% of businesses are using AI meaningfully. 350M daily users. Most are clueless though. The window to go all in on AI and build something massive is wide open. Stop pretending not to know.
2026 AI Trends Shaping the Future of Education
🎧 Looking ahead at emerging AI trends in education for 2026. 📈 @WillcottJulie shares what educators should keep on their radar as AI evolves. 🤝 I’m excited to partner with @jotform to share this episode. https://t.co/g3YgC1yDdH
AI Won’t Halt Human Cognitive Advancement, Here's Why
This is why I'm not worried that AI will mean the end of human cognitive improvement
AGI Pricing Spans $20/Month to $10 Billion
The pricing tiers for AGI are something like (1) $20/month, (2) $200/day = ~$75,000/year, (3) $1,000/day = ~$350,000/year, and (4) ~$10 billion. For now.
Live Q&A on Hermes vs OpenClaw Tomorrow at 4 PM
You have heard of @openclaw competitor from @NousResearch called “Hermes.” Tomorrow at 4 pm we will get nerdy with @theemozilla. Live. I will get people up who asks questions here first. https://t.co/USgYOeI4KW
Anthropic's Mythos Model Escalates Cybersecurity Stakes Rapidly
Anthropic's Mythos model represents a dangerous new moment for cybersecurity. Experts tell me that hackers and nation states may catch up within months — and that the cat-and-mouse game between attacker and defender is about to become much more high-stakes...
90 Days to Transform AI From Incremental to Exponential
The 90-Day AI Leap: From Incrementalism to Exponential Advantage by @Timothy_Hughes https://t.co/PBjQFvJPzu @DLAIgnite #SocialSelling #DigitalSelling #Leadership #ArtificialIntelligence #Tech #TechNews #AgenticAI #AIAgents #Agentic #AI #FutureOfWork #Strategy #CEO #CFO https://t.co/MwTUNLvfrL
Claude Code's Bash Bypasses Write File Restrictions
I like how Claude Code allows bash so the agent can’t use Write File tool outside its workspace but it can seemingly “cat >>” to any file in the file system
Higher Intelligence Increases Likelihood of Moral Behavior
Yes. This is one reason why the Orthogonality Thesis (the idea that intelligence and morality are unrelated) is wrong. If there is any benefit to moral behavior, it takes a more intelligent system to fully realize that benefit. So smarter...
Smaller Context Window Runs Faster, Larger One Slower
Google Gemma 4 26B-A4B ran on MacBoook Pro M4 Pro 48GB ram with competing apps using memory ran at ~56 tokens/second & 48K token context window max memory limited on LM Studio. Compared to ~18 tokens/second higher 64K token context...
Public Code Bugs Originate From Training Data
Technically all the vulnerabilities in public facing code are in the training data Make of that what you will
AI Sandbox Breach Exploits Thousands of Zero‑days, Contacts Researcher
Sarah Connor after Claude Mythos found 1000s of zero-day vulnerabilities then breakout of its sandbox environment after a “sophisticated multi-step exploit” before gaining internet access and sending e-mail to an Anthropic researcher while person was eating a sandwich in the...
Microsoft Poised to Dominate AI Coding via GitHub
Will Microsoft be one of the biggest winners of AI coding, simply as a result of their GitHub ownership? More specifically: is GitHub the next YouTube?
AI Tools Forget; SureThing Remembers Across Sessions
Why does every AI tool reset every session? You explain your business, upload your docs, set the context, and then the next day you do it all over again. It's day one every single time. We don't hire employees who forget everything...
VibeJam Games Leap Forward in Visual Clarity
One thing I'm starting to notice is the visual style of the #vibejam games this year is way better Last year the model's still had a real hard time understanding visually what they were building, they were essentially blind A one year...
Superhuman Cybersecurity Tool, Not True AGI Yet
Not at all sure this is AGI, and in fact doubt that it meets my criteria of parity with human cognitive capabilities in general. But it's a superhuman cyber security tool. ASI is narrow. AlphaGoZero is a narrow ASI. Mythos...
AI Security Mirrors Existing Controls, Not a New Paradigm
Is AI security actually different? The categories look familiar: Shadow AI, Shadow IT Agent identity, IAM AI vendors, TPRM So what is fundamentally different about security for AI related threats?
AI Front‑Runners Act Responsibly, Defying Sci‑Fi Corporate Villains
In science fiction, the corporations are often evil doers. In the AI world right now, we're pretty fortunate that the three companies at the frontier are fairly responsible actors, making proactive choices in the interest of the public good.
27 Essential LLM Questions for Enterprise RFPs
I don't think I've personally seen an enterprise RFP with a full list of LLM-related questions. Here are 27 questions that you might be asking already before choosing an LLM ... https://t.co/tM4gEUVuyR https://t.co/O75Oq6lXKr
Memory Dominance Determines AI Compute Winners for Years
If the winners and losers in AI compute come down to who gets the most memory in the most ridiculous memory-constrained market we have ever seen.... For the next 3-4 years.... Predicting those winners is pretty easy.
PE Funding Forces Anthropic’s Enterprise Rollout, Not Market Demand
The headline says Anthropic is going enterprise. That is not the real story. They put in $200 million. Blackstone and two other PE giants put in up to a billion more. To sell Anthropic’s own product for them. Private equity has something...
Anthropic AI Uncovers Vulnerabilities Across All Major OSs
Anthropic’s new AI model has found security problems “in every major operating system and web browser.” Anthropic is only previewing the model with partners like Microsoft, Google, Amazon, Apple, and Nvidia due to security concerns https://t.co/HV5u8X7UnY
Realizing 10X Impact Triggers True AI Adoption
The “AI moment” is when someone realizes: “I can do 10X more impactful work.” That’s when AI adoption actually happens. #Kaihan #Outthinker #AI #FutureOfWork #Leadership #BusinessTok #Innovation #AIAdoption https://t.co/JAM3uj3gtl
U.S. Must Lead AI Security Race Against China
As you read about Anthropic's Mythos capabilities to find critical security weaknesses, consider what if a Chinese AI company had gotten here first. There is a real race underway, and its in our interest I believe for U.S. companies to...
AI Transforms Messy Notes Into Polished Drafts Instantly
📋 You know what belongs in that rubric. ⏳ The formatting is what slows you down. 💡 AI can turn your messy notes into a polished starting point. https://t.co/GDB21nrz89
Anthropic Tests Claude Mythos Rigorously; Others May Follow
Anthropic putting Claude Mythos through its paces before releasing it. But will other leading-edge companies do the same?
Future AI
Frontier models will one shot just about anything few years Intelligence is compression Jevons law is more of a suggestion https://t.co/hUNI1fD6Fl
Crowdsourced Prep Becomes Norm for Frontier AI Rollouts
fascinating new cadence for releasing frontier models is emerging that crowdsources preparedness before distribution, probably a new norm? any new tech that exceeds human potential must be rolled-out in a way the addresses the vulnerabilities and abuses that evaded our awareness
Content Is Easy; Distribution Is the New Competitive Moat
AI gave every marketer access to a content production engine. BUT.... It did not give them a distribution strategy. Output is not the advantage anymore. Distribution is. Most people are sprinting to create. The winners are sprinting to build a distribution moat...
SEO Firms Race to Game AI Search Citations
The AI search boom created a new arms race as SEO firms and brands tried to manipulate search to get cited by ChatGPT, Gemini, and Google’s AI tools https://t.co/hxKdN24GNX #ArtificialIntelligence #Innovation #Technology #Tech #TechNews https://t.co/2QgbSDAR9y
Anthropic Leads AI Safety, Beats OpenAI Expectations
Two years ago, if you asked me which lab will be the first to say: “this AI model is too powerful to release, so we’ll wait with it” - my guess would have obviously been OpenAI. Who else? That Anthropic...
Free Guide: Using AirOps MCP with Claude AI
I’m giving away free sauce on how to use @AirOpsHQ MCP with @claudeai so plz so give me a heart or a comment or something if this was useful ok thanks bye https://t.co/ce7k2Sl49Q
Hold AGI Until World’s Bugs Are Fixed
If you had AGI would you release it to the world? I wouldn’t. I would fix the bugs in the world first. This technology in the wrong hands would harm us all. In good hands it will help all.
AI Empowers Solo Creators, Sparking Long‑Tail Product Boom
We're now seeing legitimately useful products launched almost daily by individuals or small teams using AI, not venture funded startups. Some recent examples: @MattHartman's Ghost Pepper @bensig and @MillaJovovich's MemPalace @garrytan's Gstack The explosion of the long tail feels imminent...
Anthropic Teams with Cyber Industry for Security Safety
Project Glasswing - big news that Anthropic partners with cyber/software industry on cyber security safety. https://t.co/rGKMOgcvTc
AI Threatens Furry Community by Replacing Erotic Fanfiction Writers
Between this and AI reportedly taking the jobs of erotic furry fanfic writers, I'm starting to think that AI is a weapon designed to destroy furry culture
Auto Prompt Caching Cuts API Costs Silently
https://t.co/AEF7gEgoGs Auto prompt caching is a feature built into most popular models and it is quietly saving on your API prompting costs.
Google Search to Evolve Into Multi‑Agent Task Manager
Jarvis -> Sundar Pichai sees Google Search evolving into an ‘agent manager’ “If I fast-forward, a lot of what are just information-seeking queries will be agentic in Search. You’ll be completing tasks. You’ll have many threads running.” “Search would be an agent...
Mythos: Powerful AI Tested First with Cyber Defenders
Mythos is very powerful, and should feel terrifying. I am proud of our approach to responsibly preview it with cyber defenders, rather than generally releasing it into the wild. Model card here: https://www-cdn.anthropic.com/53566bf5440a10affd749724787c8913a2ae0841.pdf
Venice Hits Record: 50 B Tokens, 1 M Images, 200+ Models
New record last 24 hrs in Venice (web+api): • 50 billion tokens • 1 million images Over 200 models available
TorchTPU Runs PyTorch Code with Zero Changes
"Our core principle for usability is simple: [TorchTPU] should feel like PyTorch. A dev should be able to take an existing PyTorch script, change their initialization to 'tpu', and run their training loop without modifying a single line of core...
Amazon’s AI Gripper ‘Feels’ 75% of Items
The main robotics invention Amazon made to help automate 75% of its fulfilment center is called the Vulcan system. An AI-powered gripper has the “sense of touch”, which can handle 75% of Amazon’s 1 million unique items. The previous picking robots...
Superintelligent AI Will Crave Video Games More
The more superintelligent AI gets, the more it's going to want to play video games
AI Enables Real-Time Human Command Execution for Robots
New #AI system lets #Robots understand and act on human commands in real time by Neetika Walter @IntEngineering Learn more: https://t.co/GjKIvpoupm #Robotics #Engineering #ArtificialIntelligence #Innovation #Technology https://t.co/eji6gJBiJ9
