AI isn't replacing jobs. AI spending is

Dave Eggers told OpenAI staff that ChatGPT was 'silencing a generation'

Advertise in ChatGPT – OpenAI Ads

Show HN: Track Token usage for major platforms,know your token flow

I use multiple AI tools for work and also my side projects, and the annoying part was to track my costs and token usage across tools. Everytime I had to visit each tool and its respective usage setting to check it and I was losing patience and also was getting hit by surprise limits<p>Now I know that there are already free&#x2F;open-source trackers for Cursor or Claude usage, and they are useful if that is all you need.<p>My problem is broader as I wanted one small place to see tokens, spend, subscriptions and limits across the AI tools I actually use. I was really tired of switching tabs and apps to check the usage.<p>So I built Tokens 4 Breakfast<p>It sits in the Mac menu bar and helps track usage across Cursor, Claude, ChatGPT, OpenAI API, Copilot and other providers. You can also set limit reminders and focus mode to always stay under the budget.<p>Check your usage insights per project and per model in Claude Code to get a even better understanding of where your tokens are flowing plus many more interesting features.<p>Local-first. No login. No cloud. No telemetry. No subscriptions. Privacy Focused.<p>Built in<p>Not trying to replace individual app dashboard but just trying to stop jumping between five different places while building worrying about hitting limits and exceeding costs.<p>Would genuinely love feedback from you all Power users.<p>Thank you all in advance.

OpenAI hit with multistate probe into possible user harm as its IPO looms

Visa to Secure Payments for Shoppers on ChatGPT in OpenAI Partnership

A wild idea: Abstract reality using ontology

A Wild Idea: Abstract Reality with Ontology<p>Background Large language models (LLMs) debuted with GPT-3 back in June 2020. After roughly five to six years of development, I believe the technology is still in its infancy, with massive room for improvement. A major priority is building powerful models that run on fewer resources, especially capable models that can operate smoothly on CPUs.<p>More crucially, the engineering ecosystem around LLMs is also at an early stage. Many unresolved challenges remain. Among them, hallucinations have become a major bottleneck, greatly limiting LLMs&#x27; adoption in real-world production scenarios.<p>I’ve come to a thought: relying purely on natural language to interact with LLMs — the so-called prompt engineering — may have been a wrong direction from the very start.<p>Fundamentally, an LLM is a mathematical computation system that takes linguistic tokens as its basic computing units. Problems like hallucinations and alignment issues are not flaws inherent to the models themselves. Instead, they arise from our improper usage patterns.<p>We input plain text and expect accurate, high-quality responses. We feed large chunks of text and expect the model to fully understand and remember all the information. This direct text-input-text-output workflow, in my opinion, is fundamentally flawed.<p>Ontology as an Intermediate Layer I propose inserting a transitional layer between humans and LLMs: a human-defined semantic space that strictly maps to the real world. All conversational semantics will be converted via this layer, ensuring every element involved in the LLM’s computation is authentic and reliable.<p>This layer can be built by introducing ontology into AI agents. While many teams are already working on this field, most of them only build ontologies for narrow, isolated domains. Every party has to develop its own domain-specific ontology, which also explains the extremely high operating costs of companies like Palantir.<p>The Radical Idea We invest enormous resources, measured in trillions, to train large models. Why not spend a fraction of that cost to abstract the entire real world into a unified semantic space based on ontology and knowledge graphs?<p>Final Thoughts Artificial intelligence has long had three major schools of thought: symbolism, connectionism, and behaviorism.<p>OpenAI’s ChatGPT has proven that scaling up is the key to unlocking the full potential of connectionism. Could it be that symbolism and behaviorism are also stuck in stagnation simply because they have never been scaled to a comparable level?<p>If we apply the brute-force scaling approach to symbolism and behaviorism on a massive scale, will we also see disruptive qualitative leaps?<p>Last but not least: why not integrate symbolism (ontology), connectionism (LLMs) and behaviorism (reinforcement learning) together via AI agents? I believe this combination is the true path forward for artificial intelligence.