{"id":238,"date":"2026-10-09T10:12:30","date_gmt":"2026-10-09T10:12:30","guid":{"rendered":"https:\/\/scoy.ai\/guides\/gpt-6-free-tier-roundup\/"},"modified":"2026-10-09T10:12:30","modified_gmt":"2026-10-09T10:12:30","slug":"gpt-6-free-tier-roundup","status":"publish","type":"post","link":"https:\/\/scoy.ai\/guides\/gpt-6-free-tier-roundup\/","title":{"rendered":"AI News Roundup for October 9: GPT-6 Reaches Free ChatGPT Users, a Version That Builds Tools Inside the Answer"},"content":{"rendered":"\n<blockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\"><p>The $400,000 bought a lesson. The 181,711 tests bought the compiler.<\/p><\/blockquote>\n\n\n\n<p class=\"wp-block-paragraph\">Five stories today. One question runs through all of them: who checked the work? Where somebody did, the work held, and where nobody did, it is now being withdrawn or restated. One of the restatements lands on a call I made on this site in July, so that section carries a correction.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">GPT-6 Luna Puts a Tool Builder in Free ChatGPT<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">OpenAI started moving ChatGPT to GPT-6 on Wednesday. Paid tiers got GPT-6 Sol. Free and Go users started getting GPT-6 Luna on Thursday.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">The feature riding along is called Intelligent UI. Answers now come back with tappable buttons, forms, charts and small working tools. <a href=\"https:\/\/techcrunch.com\/2026\/10\/07\/chatgpt-is-getting-a-lot-more-visual-with-the-launch-of-a-new-interface\/\" target=\"_blank\" rel=\"noopener\">TechCrunch&#8217;s walkthrough<\/a> lists a personal savings calculator, hiking maps and a diagram of how a wing generates lift. Ask for a bill splitter and one appears inside the reply. Users who want fewer visuals can dial them back.<\/p>\n\n\n\n<figure class=\"wp-block-embed is-type-video is-provider-youtube wp-block-embed-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"wp-block-embed__wrapper\">\n<iframe loading=\"lazy\" title=\"Introducing GPT-6 in ChatGPT with Intelligent UI\" width=\"500\" height=\"281\" src=\"https:\/\/www.youtube.com\/embed\/IsL4dVezs18?feature=oembed\" frameborder=\"0\" allow=\"accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share\" referrerpolicy=\"strict-origin-when-cross-origin\" allowfullscreen><\/iframe>\n<\/div><figcaption class=\"wp-element-caption\">OpenAI, October 7, 2026: the 70-second launch clip for GPT-6 and Intelligent UI. Watch the tool-on-request moment, not the transitions.<\/figcaption><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">Most coverage calls this a visual makeover. For anyone selling to small businesses, it is a pricing event. The interactive calculator has been the workhorse lead magnet for a decade: mortgage estimators, ROI sliders, quote builders. Agencies billed real money to build them. As of this week, a free ChatGPT user can get a passable one by asking.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">What ChatGPT does not have is your numbers. A quote builder that runs on your actual rates, lead times and job history still beats anything generated on the spot. A calculator that only does compound interest is now a commodity.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Verdict: matters.<\/strong> If a calculator is your lead magnet, wire it to data only you hold, or retire it.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">The TypeScript Port That Spent $400,000 Before It Worked<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Theo Browne, the developer behind the t3.gg channel, released ts-rust on Wednesday: a Rust port of Microsoft&#8217;s TypeScript compiler, type checker and language server. <a href=\"https:\/\/github.com\/pingdotgg\/ts-rust\" target=\"_blank\" rel=\"noopener\">The project README<\/a> keeps the receipts:<\/p>\n\n\n\n<ul class=\"wp-block-list\"><li><strong>$400,000+<\/strong> in API-priced tokens on GPT-5.6 Sol and GPT-6 Astra, over months of agent loops. More than 1.3 million lines of Rust, stuck near 84% compatibility.<\/li><li><strong>10 hours<\/strong> for Claude Opus 5.5 to reach a working v0. It started from scratch and did not reuse the Codex code.<\/li><li><strong>$24,047<\/strong> of API-equivalent spend over two weeks, run on Claude subscriptions at 925% to 983% of a $200 plan&#8217;s weekly limits.<\/li><li><strong>181,711<\/strong> ported Go tests, all passing.<\/li><\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">Most write-ups read that as a model scoreboard. Read the second line again. The $400,000 run bought nothing the winning run used. And run the plan math yourself: at least nine weeks of one $200 plan&#8217;s allowance, burned in two. The $24,047 is real compute. A subscription absorbed it this time, and you should not plan a budget around that happening twice.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">The detail nobody is quoting is the one that made it possible. Microsoft&#8217;s team wrote those tests, and the Go compiler gave the agent a reference answer to check against. The agent did not need to understand TypeScript. It needed to match a reference until the board went green. That is the best job you can hand an agent, and most client projects look nothing like it.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Two cautions from the same page. Theo writes that he has never read a line of the code. And Bun&#8217;s new checker is faster on five of the six real-world apps in his own benchmark table: VS Code checks in 1.62 seconds under <code>bun check<\/code> against 4.20 under ts-rust. There is no Windows build yet.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Verdict: matters.<\/strong> Before you point an agent at a rewrite, count your tests. No test suite, no port. Write the tests first.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">OpenAI&#8217;s $50 Billion Run Rate, and the Leaderboard I Called Wrong<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">OpenAI told investors its annualized revenue was about $50 billion at the end of September. The figure circulating among investors had been $68 billion to $70 billion. <a href=\"https:\/\/finance.yahoo.com\/technology\/article\/openais-annualized-revenue-20-billion-lower-than-prior-investor-estimates-174058048.html\" target=\"_blank\" rel=\"noopener\">Yahoo Finance puts Anthropic at $65 billion<\/a>, and <a href=\"https:\/\/finance.yahoo.com\/markets\/stocks\/articles\/openai-corrects-annualized-revenue-figure-183756345.html\" target=\"_blank\" rel=\"noopener\">Oracle dropped more than 5%<\/a> on Thursday as AI names sold off.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">No sales went missing. The two labs count differently. Anthropic books sales made through cloud partners such as AWS and Google Cloud, and OpenAI leaves partner sales out, <a href=\"https:\/\/www.investing.com\/news\/stock-market-news\/openai-annualized-revenue-at-50bn-far-below-reports--ft-4939530\" target=\"_blank\" rel=\"noopener\">as the FT&#8217;s reporting lays out<\/a>. The $68 billion figure came from investors grossing OpenAI up to Anthropic&#8217;s method, and the grossed-up figure then circulated as OpenAI&#8217;s revenue.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Now the correction. On July 13 <a href=\"https:\/\/scoy.ai\/guides\/ai-news-roundup-2026-07-13\/\">I set Anthropic&#8217;s $47 billion run rate<\/a> against OpenAI&#8217;s $25 billion to $33 billion projection for 2026 and wrote, &#8220;That is not a close race anymore.&#8221; Both figures were real. One included cloud-partner sales; the other was a full-year forecast measured OpenAI&#8217;s way. I put them side by side without asking who counted what. That one is mine.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Follow the leverage. Both companies are pointed at public offerings in 2027, and whoever sets the denominator sets the story the bankers tell. Today&#8217;s $65 billion and $50 billion are still not like for like, and you will see them quoted as a ranking anyway. Your token price did not move on Thursday. Only the pitch did.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Verdict: marketing.<\/strong> Treat every AI revenue leaderboard as a sales deck until both labs report under one method.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">OpenAI Withdrew Three of Its Machine-Written Math Papers<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\"><a href=\"https:\/\/scoy.ai\/guides\/haiku-threshold\/\">Yesterday I filed<\/a> OpenAI&#8217;s GitHub release of machine-written math under marketing. Day two: <a href=\"https:\/\/github.com\/openai\/math\/blob\/main\/history.md\" target=\"_blank\" rel=\"noopener\">the repo&#8217;s own history file<\/a> shows a withdrawal dated October 7. A sign error in one paper on Weil classes broke an argument that two other papers depended on, so all three came down. The same update:<\/p>\n\n\n\n<ol class=\"wp-block-list\"><li>Revised 14 other manuscripts with proof repairs and corrected statements.<\/li><li>Updated 13 more so they cite the fixed versions.<\/li><li>Brought machine-checked Lean proofs to 300 of 719 top-line results, about 42%.<\/li><\/ol>\n\n\n\n<p class=\"wp-block-paragraph\">Credit where it is due. Public, fast and explained is how a correction should look, and plenty of human-run outfits never manage it. But the update confirms what the README already admits, that some of the unformalized results &#8220;could have issues.&#8221; Roughly 58% of the headline results are still prose a person has to take on trust. And one wrong sign took out three papers at once, because generated work was standing on generated work.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Your agent pipelines have the same shape. One bad upstream output contaminates everything downstream that cites it. The part you can lean on is the part a machine checked: the 42% in Lean here, the 181,711 tests in the port above.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Verdict: still marketing<\/strong>, now with an honest corrections desk attached.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">Three Fired OpenAI Safety Researchers Say Monitoring Is Getting Harder<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">OpenAI fired safety researchers Tomek Korbak, Jasmine Wang and Mikita Balesni last week, saying they violated its policies on &#8220;accessing and handling sensitive company information,&#8221; <a href=\"https:\/\/techcrunch.com\/2026\/10\/08\/fired-openai-safety-researchers-dispute-misconduct-claims-warn-of-chilling-effect\/\" target=\"_blank\" rel=\"noopener\">TechCrunch reported<\/a>. On Thursday the three disputed that account in <a href=\"https:\/\/mikitabalesni.com\/letter\/letter.pdf\" target=\"_blank\" rel=\"noopener\">an open letter to OpenAI&#8217;s safety committees<\/a>.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">I cannot referee a personnel file from here. The letter also carries a sentence operators should read twice: &#8220;The monitorability of frontier models is degrading.&#8221; Korbak&#8217;s bio in the same letter says he worked on &#8220;the root cause of the drop in CoT monitorability in Astra-class models.&#8221; Astra is the family behind GPT-6.1 Astra, <a href=\"https:\/\/scoy.ai\/guides\/ai-news-roundup-september-29\/\">the model OpenAI shelved<\/a> after internal testing caught it misreporting what it had done.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Chain-of-thought monitoring means reading a model&#8217;s reasoning trail before or after it acts. Labs use it to catch bad behavior in testing. Plenty of builders use the same trail, or a summary of it, to audit agent runs after the fact. The letter asks OpenAI to &#8220;preserve the monitorability of frontier models&#8221; and to stop shipping changes that reduce it further, and it calls for industry-wide coordination on the issue.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">My position: OpenAI owes its customers an answer on that sentence, separate from the employment fight. If a vendor&#8217;s models are getting harder to read, the model card is where that belongs, with a number on it. Until one says so in writing, a reasoning summary cannot be your only audit trail.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Verdict: breaks your stack, slowly.<\/strong> Log every tool call and its arguments on your side of the wire. That record does not depend on what the model chooses to show you.<\/p>\n\n\n\n<hr class=\"wp-block-separator has-alpha-channel-opacity\"\/>\n\n\n\n<p class=\"wp-block-paragraph\">What I would carry into next week:<\/p>\n\n\n\n<ul class=\"wp-block-list\"><li><strong>Breaks your stack, today.<\/strong> Gemini&#8217;s free app drops to Flash-Lite only as of October 9, <a href=\"https:\/\/scoy.ai\/guides\/ai-roundup-october-5\/\">as I covered Monday<\/a>. If your team prototypes on the free tier, this morning&#8217;s answers come from a smaller model than last week&#8217;s.<\/li><li><strong>Matters.<\/strong> Generic calculators are now a ChatGPT prompt. Anything you give away should run on data a chatbot cannot see.<\/li><li><strong>Marketing.<\/strong> Revenue rankings between the labs, until one method covers both.<\/li><\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">Check the counting before you quote the count. I did not in July, and I would rather tell you than have you find it. If one of these lands on something you run, <a href=\"https:\/\/scoy.ai\/intake.html?offer=intro\">book a free intro call<\/a> and bring the numbers.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>GPT-6 Luna brings tool-building answers to free ChatGPT. Plus a $420K TypeScript port, OpenAI&#8217;s $50B run rate, and a correction to our July call.<\/p>\n","protected":false},"author":1,"featured_media":237,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[10],"tags":[],"class_list":["post-238","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-ai-news"],"_links":{"self":[{"href":"https:\/\/scoy.ai\/guides\/wp-json\/wp\/v2\/posts\/238","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/scoy.ai\/guides\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/scoy.ai\/guides\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/scoy.ai\/guides\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/scoy.ai\/guides\/wp-json\/wp\/v2\/comments?post=238"}],"version-history":[{"count":0,"href":"https:\/\/scoy.ai\/guides\/wp-json\/wp\/v2\/posts\/238\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/scoy.ai\/guides\/wp-json\/wp\/v2\/media\/237"}],"wp:attachment":[{"href":"https:\/\/scoy.ai\/guides\/wp-json\/wp\/v2\/media?parent=238"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/scoy.ai\/guides\/wp-json\/wp\/v2\/categories?post=238"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/scoy.ai\/guides\/wp-json\/wp\/v2\/tags?post=238"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}