New AI releases with our call, security incidents, score, price and verdict moves, and guide updates, with the reason and the source.https://hayzlabs.comhttps://hayzlabs.com/reviews/changes/#n-2026-10-23-how-the-litellm-hijack-happenedhttps://hayzlabs.com/reviews/changes/#n-2026-10-23-how-the-litellm-hijack-happenedBerriAI says a master key leaked in May and never revoked let attackers publish the bad 1.107.0. If you run LiteLLM, rotate every key it held and check which image digest you run if you pull latest or main-stable. Source: BerriAI's incident report (https://docs.litellm.ai/blog/security-incident-oct-21-2026-attack-chain), The Hacker News (https://thehackernews.com/2026/10/leaked-litellm-master-key-let-attackers.html).Fri, 23 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#n-2026-10-23-minimax-m3-1-flash-weights-are-outhttps://hayzlabs.com/reviews/changes/#n-2026-10-23-minimax-m3-1-flash-weights-are-outMiniMax published the weights for its cheap agent model. Free for non-commercial use; businesses must show a 'Built with' notice, and firms over $20M a year in revenue need written permission. Best for: Cheap coding and office agents you can host yourself. You give up: Custom license with limits on business use; very large to self-host. Source: MiniMax's post (https://www.minimax.io/blog/minimax-m31f), TechNode (https://technode.com/2026/10/23/minimax-releases-open-weights-for-428-billion-parameter-m3-1-flash-model/).Fri, 23 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#n-2026-10-23-openai-switched-off-gpt-4-o1-o3-mini-and-morehttps://hayzlabs.com/reviews/changes/#n-2026-10-23-openai-switched-off-gpt-4-o1-o3-mini-and-moreOpenAI's API stopped serving older models on 23 Oct, including GPT-3.5 Turbo, GPT-4, GPT-4 Turbo, o1, o1-pro, o3-mini, o4-mini, GPT-4.1 nano and gpt-image-1. Calls to them now fail. Source: OpenAI's deprecations page (https://developers.openai.com/api/docs/deprecations).Fri, 23 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#n-2026-10-23-kimi-k3-1-prohttps://hayzlabs.com/reviews/changes/#n-2026-10-23-kimi-k3-1-proMoonshot's new top model, cheaper than Kimi K3 at $2 in and $10 out per million tokens, with limited free use on the web. Open weights are promised by 4 Nov. Best for: Long coding agents and office tasks at a lower price than frontier models. You give up: Weights are only a promise until 4 Nov; most test results are Kimi's own. Source: Moonshot's post (https://www.kimi.ai/blog/kimi-k3-1-pro), Kimi price page (https://www.kimi.ai/resources/kimi-k3-1-pro-pricing).Fri, 23 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#n-2026-10-22-openai-decisions-api-is-generally-availablehttps://hayzlabs.com/reviews/changes/#n-2026-10-22-openai-decisions-api-is-generally-availableOpenAI's endpoint for yes/no, choice and score answers left beta. It runs on GPT-6 Luna and now takes up to 400 questions per request, up from 200. Best for: High-volume routing, scoring and classification steps inside agents. You give up: Tied to GPT-6 Luna; gives decisions only, never free text. Source: OpenAI API changelog (https://developers.openai.com/api/docs/changelog).Thu, 22 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#n-2026-10-22-1password-for-agentshttps://hayzlabs.com/reviews/changes/#n-2026-10-22-1password-for-agentsLets AI agents sign in for you after you approve on your own device, without the agent seeing the password. Most named agents, such as Claude Duet, are listed as coming soon. Source: 1Password's announcement (https://1password.com/press/2026/october/1password-for-agents-trust-layer-personal-ai-agents), TechCrunch (https://techcrunch.com/2026/10/22/1password-launches-a-way-for-ai-agents-to-log-in-on-your-behalf-without-seeing-your-password/).Thu, 22 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#n-2026-10-22-olmoearth-v1-3https://hayzlabs.com/reviews/changes/#n-2026-10-22-olmoearth-v1-3Open satellite embeddings for the whole planet. Useful for map and crop research, but outside the everyday jobs we cover. Source: Ai2's OlmoEarth page (https://allenai.org/olmoearth), SiliconANGLE (https://siliconangle.com/2026/10/22/ai2-releases-olmoearth-v1-3-state-of-the-art-ai-ready-satellite-embeddings-for-the-whole-planet/).Thu, 22 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#n-2026-10-22-canva-crafthttps://hayzlabs.com/reviews/changes/#n-2026-10-22-canva-craftCanva's first image model, for every Canva user in Dream Lab. It plans layout and text before drawing. Quality claims are Canva's own so far. Best for: Canva users making posters, slides and social posts with text in the image. You give up: Only inside Canva or its API; quality claims are Canva's own so far. Source: Canva's announcement (https://www.canva.com/newsroom/news/canva-craft/), TechCrunch (https://techcrunch.com/2026/10/22/canva-releases-its-first-image-generation-model/).Thu, 22 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#n-2026-10-21-claude-api-version-2026-10-21-breaks-old-codehttps://hayzlabs.com/reviews/changes/#n-2026-10-21-claude-api-version-2026-10-21-breaks-old-codeIf you build on the Claude API: the new API version and SDK majors reject temperature, top_p and top_k, and send keys as a Bearer header. Read the migration guide before upgrading. Source: Anthropic's migration guide (https://platform.claude.com/docs/en/api/migration-2026-10-21), Claude release notes (https://platform.claude.com/docs/en/release-notes/overview).Wed, 21 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#n-2026-10-21-nemotron-3-5-embed-8bhttps://hayzlabs.com/reviews/changes/#n-2026-10-21-nemotron-3-5-embed-8bOpen embedding model, free for commercial use, that searches text and page images such as tables and scans together. A 1B version and a reranker came out alongside it. Best for: Search over scanned documents, tables and charts in your own retrieval setup. You give up: Needs a large GPU; quality claims are NVIDIA's own so far. Source: NVIDIA's post (https://huggingface.co/blog/nvidia/nemotron-retriever-world-class-agentic-retrieval), Model card (https://huggingface.co/nvidia/Nemotron-3.5-Embed-8B-BF16).Wed, 21 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#n-2026-10-21-litellm-1-107-0-was-malwarehttps://hayzlabs.com/reviews/changes/#n-2026-10-21-litellm-1-107-0-was-malwareA hijacked release on PyPI and Docker Hub stole credentials for about two hours. Pin 1.106.2. If you installed it, rebuild the machine and rotate every key it could reach. Source: LiteLLM's incident post (https://docs.litellm.ai/blog/security-incident-oct-21-2026), The Hacker News (https://thehackernews.com/2026/10/litellm-compromised-in-supply-chain.html).Wed, 21 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#n-2026-10-21-flinthttps://hayzlabs.com/reviews/changes/#n-2026-10-21-flintFree open coding model (Apache-2.0) that runs on a laptop. Speed and test figures are the maker's own; Wired's hands-on was positive. Best for: Local coding agents on a 48GB Mac or GPU with no API fees. You give up: Needs smaller and clearer tasks than paid assistants. Source: Recursive's post (https://recursive.com/updates/introducing-recursive-flint), TechCrunch (https://techcrunch.com/2026/10/21/recursive-launches-flint-a-laptop-coding-model-it-says-was-largely-built-by-ai/).Wed, 21 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#n-2026-10-21-mammoth-1-and-elephant-1https://hayzlabs.com/reviews/changes/#n-2026-10-21-mammoth-1-and-elephant-1New open "decision models". No named license yet, and the rankings come from Baseten without saying who ran them. We look again when an outside board lists them. Source: Baseten's post (https://www.baseten.co/blog/the-best-decision-models-in-the-world-mammoth-1-and-elephant-1/).Wed, 21 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#n-2026-10-21-kumo-relationalhttps://hayzlabs.com/reviews/changes/#n-2026-10-21-kumo-relationalReal and open, but a prediction tool for company databases, outside the jobs we cover. The only test results are the authors' own. Source: Hugging Face blog (https://huggingface.co/blog/nvidia/kumo-relational).Wed, 21 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#n-2026-10-20-voxtral-3https://hayzlabs.com/reviews/changes/#n-2026-10-20-voxtral-3Mistral's new speech family: transcription, text to speech and a voice agent kit from one EU-hosted vendor. Best for: Voice agents that want listening and speaking from one EU vendor. You give up: Weights and self-host license not public. Source: Mistral's post (https://mistral.ai/news/voxtral-3/), The Decoder (https://the-decoder.com/mistral-ai-launches-voxtral-3-a-full-voice-agent-stack-that-listens-thinks-and-talks-in-real-time/).Tue, 20 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#n-2026-10-20-eleven-realtime-v1https://hayzlabs.com/reviews/changes/#n-2026-10-20-eleven-realtime-v1ElevenLabs' first speech to speech models for live voice agents, in 90+ languages. $0.08 a minute plus model tokens. Best for: Live voice agents that need many languages and voices. You give up: Per-minute fee plus tokens adds up. Source: ElevenLabs' post (https://elevenlabs.io/blog/realtime), ElevenLabs' price page (https://elevenlabs.io/pricing/api).Tue, 20 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#n-2026-10-20-ming-3-0-ttshttps://hayzlabs.com/reviews/changes/#n-2026-10-20-ming-3-0-ttsOpen text to speech under MIT, with voice cloning and tags for laughs and sighs. Free for commercial use. Best for: Self-hosted voice with cloning, free for commercial use. You give up: Ranks below Fish Audio Drama 3; no memory figure. Source: Model card (https://huggingface.co/inclusionAI/Ming-3.0-tts-tiny), TechNode (https://technode.com/2026/10/20/ant-group-open-sources-ming-3-0-speech-models-with-voice-design-and-audio-understanding/).Tue, 20 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#n-2026-10-20-deepinferhttps://hayzlabs.com/reviews/changes/#n-2026-10-20-deepinferBig news, but it runs one DeepSeek model on two chip types. vLLM still does more for most teams. Source: GitHub (https://github.com/deepseek-ai/DeepInfer), Bloomberg (https://www.bloomberg.com/news/articles/2026-10-20/deepseek-unveils-open-source-tool-to-help-cut-ai-operating-costs).Tue, 20 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#n-2026-10-20-torchtpuhttps://hayzlabs.com/reviews/changes/#n-2026-10-20-torchtpuThe code went public, but it was announced in April and only helps teams already on Google Cloud TPUs. Source: GitHub (https://github.com/google-pytorch/torch_tpu).Tue, 20 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#c-2026-10-20-gemini-4-argonhttps://hayzlabs.com/reviews/changes/#c-2026-10-20-gemini-4-argonScored: now in Google AI Plus, Pro and Ultra and on the paid API. Writing 99, research 97, coding 64 and 6th on the Agent board, from Arena's open dataset; votes are still early. Verdict wait to use-it.Tue, 20 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#n-2026-10-19-minimax-m3-1-flashhttps://hayzlabs.com/reviews/changes/#n-2026-10-19-minimax-m3-1-flashA cheap fast model ($0.15 in, $0.60 out per million tokens) with a free version on OpenRouter. Not on a public board yet, so no score. Best for: High-volume simple tasks on a tight budget. You give up: VentureBeat reports a high rate of made-up answers. Source: MiniMax's post (https://www.minimax.io/blog/minimax-m31f), VentureBeat (https://venturebeat.com/technology/minimaxs-new-m3-1-flash-beats-deepseek-and-opus-5-on-a-key-business-benchmark-at-lower-cost-but-with-a-high-hallucination-rate).Mon, 19 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#n-2026-10-19-mercury-3https://hayzlabs.com/reviews/changes/#n-2026-10-19-mercury-3Cheap and fast on paper ($0.20 in, $0.70 out per million tokens, half price at launch), but every quality claim is Inception's own. We look again when an outside board lists it. Source: Inception's post (https://www.inceptionlabs.ai/blog/introducing-mercury-3).Mon, 19 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#n-2026-10-19-band-desktophttps://hayzlabs.com/reviews/changes/#n-2026-10-19-band-desktopLaunched 15 Oct with no price, no access details and no outside tests, so there is nothing yet to judge it on. Source: BAND docs (https://docs.band.ai/band-desktop).Mon, 19 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#n-2026-10-19-flux-3-maxhttps://hayzlabs.com/reviews/changes/#n-2026-10-19-flux-3-maxFast text and image to video built on FLUX 3 Video, at a launch discount until 7 Nov. Speed claims are fal's own, and only one major outlet covered it. Source: fal's page (https://fal.ai/flux-3-max), VentureBeat (https://venturebeat.com/technology/fal-launches-flux-3-max-claiming-22-faster-ai-video-generation-than-seedance-2-5).Mon, 19 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/ai-at-the-doctorshttps://hayzlabs.com/guides/ai-at-the-doctorsNew guide: AI at the doctor's: the listening scribe, AI-drafted replies, and checking your visit note.Mon, 19 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/ai-glasseshttps://hayzlabs.com/guides/ai-glassesNew guide: AI glasses: what do they record, which settings matter, and what can I do if I'm filmed?.Mon, 19 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/before-you-share-an-app-you-madehttps://hayzlabs.com/guides/before-you-share-an-app-you-madeNew guide: Is the app I made with AI safe to share? A 20-minute check before you send the link.Mon, 19 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/connect-medical-records-to-aihttps://hayzlabs.com/guides/connect-medical-records-to-aiNew guide: Connect your medical records to an AI? A decision and a setup check.Mon, 19 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/fake-intimate-image-first-48-hourshttps://hayzlabs.com/guides/fake-intimate-image-first-48-hoursNew guide: Someone made a fake nude or sexual image of me (or my child): what do I do in the first 48 hours?.Mon, 19 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/interviewed-by-aihttps://hayzlabs.com/guides/interviewed-by-aiNew guide: Interviewed or screened by AI: what to expect, how to prepare, how to ask for a person.Mon, 19 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/is-ai-coming-for-my-jobhttps://hayzlabs.com/guides/is-ai-coming-for-my-jobNew guide: Is AI coming for my job? A 30-minute task-by-task check and a skills plan.Mon, 19 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/switch-assistants-take-your-historyhttps://hayzlabs.com/guides/switch-assistants-take-your-historyNew guide: Switching AI assistants: how do I take my history with me?.Mon, 19 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/teaching-with-aihttps://hayzlabs.com/guides/teaching-with-aiNew guide: Teaching with AI: how do I plan and give feedback without breaking my school's rules?.Mon, 19 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/when-an-ai-answers-the-phonehttps://hayzlabs.com/guides/when-an-ai-answers-the-phoneNew guide: An AI answered the phone: how do I reach a person, and should I let an AI call for me?.Mon, 19 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#n-2026-10-13-gemini-4-argon-opens-uphttps://hayzlabs.com/reviews/changes/#n-2026-10-13-gemini-4-argon-opens-upGoogle's top model is now on the paid Gemini API and in Google AI Plus from $4.99 a month. It leads Arena's text votes; votes are still early. Best for: Everyday chat and writing near the top of Arena votes, from $4.99 a month. You give up: Behind Claude and GPT for web app coding; API price doubles on 1 Dec. Source: Google's plans page (https://gemini.google/subscriptions/), Gemini API pricing (https://ai.google.dev/gemini-api/docs/pricing).Tue, 13 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#c-2026-10-11-chatgpt-gpt-6https://hayzlabs.com/reviews/changes/#c-2026-10-11-chatgpt-gpt-6Audit under our rule that each plan is rated on the best model it includes: OpenAI names GPT-6 Sol for Plus chat and GPT-6 Luna for Free, so both checklist lines now use those models instead of the best model in Plus. Coding fell because GPT-6 Sol ranks lower after the Text Coding check, so Free now scores higher. Research and writing keep their ratings but now come from Free. Dropped the Meta Muse test and Instruction Following board, and stopped counting the 24 Sep Tom's Guide test, which did not name the ChatGPT-6 version.Sun, 11 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#c-2026-10-11-chatgpt-gpt-6-2https://hayzlabs.com/reviews/changes/#c-2026-10-11-chatgpt-gpt-6-2Rule change: each plan is scored on the best model it lets you choose. OpenAI's help page says Plus includes GPT-6 Astra in ChatGPT Work and Codex, while Plus chat offers only GPT-6 Sol. Agents and coding on Plus are now rated on Astra, so coding returns to 7/10. Research and writing stay on chat models, so they are unchanged. Arena text URLs moved to the /leaderboard/chat/text/ form.Sun, 11 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#c-2026-10-11-claudehttps://hayzlabs.com/reviews/changes/#c-2026-10-11-claudeAudit under our rule that each plan is rated on the best model it includes: Anthropic names no default for Pro, so Pro is now rated on its best model for each job instead of an assumed Sonnet 5.5. Writing on Pro rose to 7/10 on Opus 5.5. Other ratings unchanged. Dropped the 27 Aug Tom's Guide test, which predates the current models, and stopped counting the Opus 5.5 test toward the Free plan, which runs Sonnet 5.5.Sun, 11 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#c-2026-10-11-claude-codehttps://hayzlabs.com/reviews/changes/#c-2026-10-11-claude-codeRated for coding by the checklist. Artificial Analysis's methodology page shows v1.5 has been in use since September 2026 and is the first version with Terminal-Bench 4.0, so its scores fall inside the 90-day window. Agents not rated yet.Sun, 11 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#c-2026-10-11-claude-fable-5-1https://hayzlabs.com/reviews/changes/#c-2026-10-11-claude-fable-5-1New tie rule: models tied on the Artificial Analysis index take the lowest rank in the tie, so research is rated on that.Sun, 11 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#c-2026-10-11-claude-sonnet-5-5https://hayzlabs.com/reviews/changes/#c-2026-10-11-claude-sonnet-5-5Audit: re-derived the ratings with our rule of counting each model once on a leaderboard. Coding moved down a step (Text Coding is 16th of distinct models, below its WebDev rank). Writing moved up a step (Instruction Following is 10th of distinct models). best_category now agents. Agent cost relabelled as the 14-day median; PCWorld cited for the free-plan model.Sun, 11 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#c-2026-10-11-gemini-apphttps://hayzlabs.com/reviews/changes/#c-2026-10-11-gemini-appAudit under our rule that each plan is rated on the best model it includes: Google names no default for AI Pro, so AI Pro is now rated on its better model, 3.8 Flash, instead of the lower-ranked 3.1 Pro. Writing rose to 7/10. Research, coding and agents unchanged. Verdict now depends: it scores best for writing, where Claude Pro scores the same, so Gemini is not outclassed there.Sun, 11 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#c-2026-10-11-mistral-large-4https://hayzlabs.com/reviews/changes/#c-2026-10-11-mistral-large-4Rated: it is publicly available as an API preview and its arena.ai rows are not marked preliminary. Coding and agents scored highest; research (moved up a step by the Artificial Analysis index) and writing (moved up a step by Instruction Following) scored lower. Verdict stays wait: weights are due 31 Oct and Mistral says training continues.Sun, 11 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#c-2026-10-11-openai-codex-clihttps://hayzlabs.com/reviews/changes/#c-2026-10-11-openai-codex-cliRated for coding by the checklist. Artificial Analysis's methodology page shows v1.5 has been in use since September 2026, so its scores fall inside the 90-day window. Agents not rated yet.Sun, 11 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#c-2026-10-11-opencodehttps://hayzlabs.com/reviews/changes/#c-2026-10-11-opencodeRated for coding by the checklist. Artificial Analysis's methodology page shows v1.5 has been in use since September 2026, so its score falls inside the 90-day window. Agents not rated yet.Sun, 11 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#log-2026-10-11-plain-wordshttps://hayzlabs.com/reviews/changes/#log-2026-10-11-plain-wordsBeginner and Intermediate pages now explain terms such as API, token and context window the first time they appear.Sun, 11 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#log-2026-10-11-codehttps://hayzlabs.com/reviews/changes/#log-2026-10-11-codeCode on the site was re-read and fixed where needed, and scripts we could not test now say so.Sun, 11 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#log-2026-10-11-factshttps://hayzlabs.com/reviews/changes/#log-2026-10-11-factsFacts that change over time, such as plan limits, laws, scam figures and vendor features, were re-checked against their sources on 11 October. Eight were corrected or dated.Sun, 11 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#log-2026-10-11-hands-onhttps://hayzlabs.com/reviews/changes/#log-2026-10-11-hands-onInteractive pieces are now called Hands-on tools, on the home page and the Tools page.Sun, 11 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#log-2026-10-11-removed-searchhttps://hayzlabs.com/reviews/changes/#log-2026-10-11-removed-searchRemoved the hands-on tool on Answers or search results, which only repeated the page's own table.Sun, 11 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#log-2026-10-11-removed-when-nothttps://hayzlabs.com/reviews/changes/#log-2026-10-11-removed-when-notRemoved the hands-on tool on When not to use AI, which only repeated the page's own table.Sun, 11 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#log-2026-10-11-boxhttps://hayzlabs.com/reviews/changes/#log-2026-10-11-boxThe box at the top of every page now has three plain parts: What this is, Why it matters and The short version.Sun, 11 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#log-2026-10-11-follow-alonghttps://hayzlabs.com/reviews/changes/#log-2026-10-11-follow-alongWe followed each Beginner and Intermediate page step by step, updated settings paths that had moved and added steps that were missing.Sun, 11 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/ads-in-ai-answershttps://hayzlabs.com/guides/ads-in-ai-answersNew guide: Ads and sponsored answers in AI chats: how to spot them and get a straight recommendation.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/agents-near-your-email-and-fileshttps://hayzlabs.com/guides/agents-near-your-email-and-filesNew guide: Letting an agent near your email, browser and files.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/ai-built-into-your-appshttps://hayzlabs.com/guides/ai-built-into-your-appsNew guide: The AI already in your phone, email and office apps: what it reads, and how to turn it up or down.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/ai-detectorshttps://hayzlabs.com/guides/ai-detectorsNew guide: AI detectors: what they can and can't tell you.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/ai-memory-audithttps://hayzlabs.com/guides/ai-memory-auditNew guide: AI memory: what it keeps, and a ten-minute audit.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/ai-phishing-and-passkeyshttps://hayzlabs.com/guides/ai-phishing-and-passkeysNew guide: AI-written phishing: stop looking for typos, set up passkeys.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/tools/am-i-actually-fasterhttps://hayzlabs.com/tools/am-i-actually-fasterNew tool: Am I actually faster?.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/answers-or-search-resultshttps://hayzlabs.com/guides/answers-or-search-resultsNew guide: AI answers or search results: when to use which.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/workflows/asking-questions-of-your-datahttps://hayzlabs.com/workflows/asking-questions-of-your-dataNew workflow: Asking questions of your data when you're not an analyst.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/build-a-small-agenthttps://hayzlabs.com/guides/build-a-small-agentNew guide: Build a small agent yourself: one tool, a step limit and a log.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/build-an-assistant-others-will-usehttps://hayzlabs.com/guides/build-an-assistant-others-will-useNew guide: Build an assistant other people will use.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/workflows/check-a-claim-you-saw-onlinehttps://hayzlabs.com/workflows/check-a-claim-you-saw-onlineNew workflow: Check a claim you saw online, with AI as a helper, not the judge.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/check-an-image-or-videohttps://hayzlabs.com/guides/check-an-image-or-videoNew guide: Was this image made or changed by AI? A five-minute check.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/start/check-before-you-trusthttps://hayzlabs.com/start/check-before-you-trustNew start here: Check before you trust: a five-minute routine.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/workflows/checking-agent-workhttps://hayzlabs.com/workflows/checking-agent-workNew workflow: Checking agent work: review loops and work you didn't watch.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/coding-agents-for-non-codershttps://hayzlabs.com/guides/coding-agents-for-non-codersNew guide: Coding agents for people who don't write code.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/college-applications-and-aihttps://hayzlabs.com/guides/college-applications-and-aiNew guide: College and university applications: what AI use is allowed, and how to use it honestly.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/workflows/complain-and-get-it-fixedhttps://hayzlabs.com/workflows/complain-and-get-it-fixedNew workflow: Complain about a faulty product or poor service, with AI drafting and you checking your rights.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/describe-the-failurehttps://hayzlabs.com/guides/describe-the-failureNew guide: Describe the failure: reports an agent can act on.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/setup/do-i-need-to-pay-for-aihttps://hayzlabs.com/setup/do-i-need-to-pay-for-aiNew setup: Do I need to pay for AI, and for which plan?.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/first-api-call-with-a-spending-caphttps://hayzlabs.com/guides/first-api-call-with-a-spending-capNew guide: Your first API call, with a spending cap.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/workflows/fix-a-tech-problem-with-aihttps://hayzlabs.com/workflows/fix-a-tech-problem-with-aiNew workflow: Fix a computer, phone or app problem with AI, safely.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/getting-disagreement-instead-of-agreementhttps://hayzlabs.com/guides/getting-disagreement-instead-of-agreementNew guide: Getting disagreement instead of agreement.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/workflows/group-a-list-by-meaninghttps://hayzlabs.com/workflows/group-a-list-by-meaningNew workflow: Find near-duplicates and group a long list by meaning (embeddings).Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/hang-up-and-call-backhttps://hayzlabs.com/guides/hang-up-and-call-backNew guide: AI voice scams: hang up and call back on a number you know.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/health-money-and-legal-questionshttps://hayzlabs.com/guides/health-money-and-legal-questionsNew guide: Health, money and legal questions: green lights and red lines.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/help-an-older-relative-start-with-aihttps://hayzlabs.com/guides/help-an-older-relative-start-with-aiNew guide: Help an older relative start using AI, safely.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/workflows/inbox-letters-and-billshttps://hayzlabs.com/workflows/inbox-letters-and-billsNew workflow: Personal admin with AI: inbox, letters and bills, with you approving every send and payment.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/workflows/job-search-and-interview-prephttps://hayzlabs.com/workflows/job-search-and-interview-prepNew workflow: Job search and interview prep.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/kids-homework-and-a-family-ai-planhttps://hayzlabs.com/guides/kids-homework-and-a-family-ai-planNew guide: Kids, homework and a family AI plan.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/learn-to-code-with-aihttps://hayzlabs.com/guides/learn-to-code-with-aiNew guide: Learn to code with AI without skipping the learning.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/learning-without-losing-the-learninghttps://hayzlabs.com/guides/learning-without-losing-the-learningNew guide: Learning with AI without losing the learning.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/local-or-cloudhttps://hayzlabs.com/guides/local-or-cloudNew guide: Local or cloud: when a model on your own device is enough.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/workflows/long-agent-jobshttps://hayzlabs.com/workflows/long-agent-jobsNew workflow: Long agent jobs: pieces, milestones and stop rules.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/make-an-image-with-aihttps://hayzlabs.com/guides/make-an-image-with-aiNew guide: Make an image with AI: describe it well, check it, and know what you can use it for.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/workflows/meeting-notes-to-decisionshttps://hayzlabs.com/workflows/meeting-notes-to-decisionsNew workflow: Meeting notes to decisions and owners.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/workflows/one-prompt-many-rowshttps://hayzlabs.com/workflows/one-prompt-many-rowsNew workflow: Run one prompt over hundreds of rows, then measure how often it's wrong.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/open-source-ai-toolshttps://hayzlabs.com/guides/open-source-ai-toolsNew guide: Open-source AI tools: one pick for each job.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/workflows/payment-change-checkhttps://hayzlabs.com/workflows/payment-change-checkNew workflow: Check the request, not the voice: payment changes at work.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/workflows/personal-eval-sethttps://hayzlabs.com/workflows/personal-eval-setNew workflow: Test any new model on your own tasks (a personal eval set).Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/workflows/photos-in-data-outhttps://hayzlabs.com/workflows/photos-in-data-outNew workflow: Photos in, data out: forms, receipts, whiteboards.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/workflows/plan-a-trip-then-verifyhttps://hayzlabs.com/workflows/plan-a-trip-then-verifyNew workflow: Plan a trip, then verify every stop.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/workflows/plan-meals-with-aihttps://hayzlabs.com/workflows/plan-meals-with-aiNew workflow: Plan a week of meals and a shopping list with AI, and check what matters.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/point-your-phone-at-ithttps://hayzlabs.com/guides/point-your-phone-at-itNew guide: Point your phone at it: what AI can and can't tell you from a photo.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/workflows/practise-a-language-with-aihttps://hayzlabs.com/workflows/practise-a-language-with-aiNew workflow: Practise a new language with AI: speak daily, get corrected, track progress.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/workflows/prepare-a-negotiation-with-aihttps://hayzlabs.com/workflows/prepare-a-negotiation-with-aiNew workflow: Prepare for a negotiation with AI: a pay rise, a job offer or a price.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/workflows/presentation-with-aihttps://hayzlabs.com/workflows/presentation-with-aiNew workflow: Make a presentation with AI: your outline first, then check every slide.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/setup/privacy-settings-passhttps://hayzlabs.com/setup/privacy-settings-passNew setup: Privacy settings pass: 15 minutes, once a quarter.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/prompt-retrieve-or-fine-tunehttps://hayzlabs.com/guides/prompt-retrieve-or-fine-tuneNew guide: Prompt, retrieve or fine-tune? Getting a model to use your material.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/start/prompting-in-one-pagehttps://hayzlabs.com/start/prompting-in-one-pageNew start here: Prompting in one page.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/workflows/read-a-contract-with-aihttps://hayzlabs.com/workflows/read-a-contract-with-aiNew workflow: Read a contract, lease or terms with AI before you sign.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/workflows/read-the-sourcehttps://hayzlabs.com/workflows/read-the-sourceNew workflow: Read the source, use the summary as a map.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/workflows/reconcile-two-exportshttps://hayzlabs.com/workflows/reconcile-two-exportsNew workflow: Reconcile two exports.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/workflows/research-with-citation-checkinghttps://hayzlabs.com/workflows/research-with-citation-checkingNew workflow: Research a topic with citation checking.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/script-workflow-or-agenthttps://hayzlabs.com/guides/script-workflow-or-agentNew guide: Script, workflow or agent: how much to automate.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/second-language-and-dyslexiahttps://hayzlabs.com/guides/second-language-and-dyslexiaNew guide: Writing in a second language, or with dyslexia.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/workflows/spreadsheet-formula-then-testhttps://hayzlabs.com/workflows/spreadsheet-formula-then-testNew workflow: Get a spreadsheet formula from AI, then prove it works.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/start/start-herehttps://hayzlabs.com/start/start-hereNew start here: Start here.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/structured-output-you-can-trusthttps://hayzlabs.com/guides/structured-output-you-can-trustNew guide: Get data you can trust from an API: schemas, validation and retries.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/talk-then-edithttps://hayzlabs.com/guides/talk-then-editNew guide: Talk, then edit: dictation and voice mode.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/talking-to-ai-when-strugglinghttps://hayzlabs.com/guides/talking-to-ai-when-strugglingNew guide: Talking to AI when you're struggling: what it can help with, and where it stops.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/taxes-with-aihttps://hayzlabs.com/guides/taxes-with-aiNew guide: Taxes with AI: what it can help with, and what you check on the official site.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/start/tell-it-who-you-arehttps://hayzlabs.com/start/tell-it-who-you-areNew start here: Tell it who you are, once.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/workflows/transcribe-then-checkhttps://hayzlabs.com/workflows/transcribe-then-checkNew workflow: Transcribe a recording, then check it before you quote it.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/translating-with-aihttps://hayzlabs.com/guides/translating-with-aiNew guide: Translating with AI: when it's good enough, and how to check.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/start/what-not-to-pastehttps://hayzlabs.com/start/what-not-to-pasteNew start here: What not to paste.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/guides/when-not-to-use-aihttps://hayzlabs.com/guides/when-not-to-use-aiNew guide: When not to use AI.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/tools/where-could-ai-help-this-weekhttps://hayzlabs.com/tools/where-could-ai-help-this-weekNew tool: Where could AI help this week?.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/workflows/write-a-speech-with-aihttps://hayzlabs.com/workflows/write-a-speech-with-aiNew workflow: Write a speech, toast or eulogy with AI that still sounds like you.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/workflows/writing-brief-draft-checkhttps://hayzlabs.com/workflows/writing-brief-draft-checkNew workflow: Writing: brief, draft, check.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/start/your-first-useful-hourhttps://hayzlabs.com/start/your-first-useful-hourNew start here: Your first useful hour: three everyday tasks.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#log-2026-10-10-merge-prompt-libraryhttps://hayzlabs.com/reviews/changes/#log-2026-10-10-merge-prompt-libraryA prompt library is now a section of Prompting in one page. The old address redirects there.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#log-2026-10-10-copyhttps://hayzlabs.com/reviews/changes/#log-2026-10-10-copyCode blocks have a Copy button, and blocks that hold a whole file have a Download button.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#log-2026-10-10-levelshttps://hayzlabs.com/reviews/changes/#log-2026-10-10-levelsEvery page now opens with an In short box and carries a level: Beginner, Intermediate or Advanced.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#log-2026-10-10-qualityhttps://hayzlabs.com/reviews/changes/#log-2026-10-10-qualityFigures across the site were rechecked against their sources, and the weakest pages were rewritten with clearer steps.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#log-2026-10-10-x-baseratehttps://hayzlabs.com/reviews/changes/#log-2026-10-10-x-baserateHands-on tool on AI detectors: see how many flagged essays are false alarms on a grid of 1,000.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#log-2026-10-10-x-callhttps://hayzlabs.com/reviews/changes/#log-2026-10-10-x-callHands-on tool on AI voice scams: practise spotting the messages that mean hang up and call back.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#log-2026-10-10-x-lookalikehttps://hayzlabs.com/reviews/changes/#log-2026-10-10-x-lookalikeHands-on tool on AI-written phishing: spot the lookalike links.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#log-2026-10-10-x-agentloophttps://hayzlabs.com/reviews/changes/#log-2026-10-10-x-agentloopHands-on tool on Build a small agent: step through an agent run and watch the tokens add up.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#log-2026-10-10-x-filldownhttps://hayzlabs.com/reviews/changes/#log-2026-10-10-x-filldownHands-on tool on Get a spreadsheet formula from AI: fill a column down and see what each kind of cell reference does.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#log-2026-10-10-x-replyhttps://hayzlabs.com/reviews/changes/#log-2026-10-10-x-replyHands-on tool on Get data you can trust from an API: edit a model reply and watch the parse, schema and value checks.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#log-2026-10-10-x-thresholdhttps://hayzlabs.com/reviews/changes/#log-2026-10-10-x-thresholdHands-on tool on Group a list by meaning: move the similarity threshold and see which hand-checked pairs it gets right.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#log-2026-10-10-x-lighthttps://hayzlabs.com/reviews/changes/#log-2026-10-10-x-lightHands-on tool on Health, money and legal questions: sort requests by the green, amber and red table.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#log-2026-10-10-x-chainhttps://hayzlabs.com/reviews/changes/#log-2026-10-10-x-chainHands-on tool on Long agent jobs: see how the odds of a clean run fall as steps add up.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#log-2026-10-10-x-failurehttps://hayzlabs.com/reviews/changes/#log-2026-10-10-x-failureHands-on tool on Prompt, retrieve or fine-tune: sort failures into missing knowledge or wrong behaviour.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#log-2026-10-10-x-briefhttps://hayzlabs.com/reviews/changes/#log-2026-10-10-x-briefHands-on tool on Prompting in one page: build a brief one part at a time.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#log-2026-10-10-x-matchhttps://hayzlabs.com/reviews/changes/#log-2026-10-10-x-matchHands-on tool on Reconcile two exports: watch the script match rows in order.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#log-2026-10-10-x-rungshttps://hayzlabs.com/reviews/changes/#log-2026-10-10-x-rungsHands-on tool on Script, workflow or agent: answer the five questions and see which rung fits.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#log-2026-10-10-x-switchhttps://hayzlabs.com/reviews/changes/#log-2026-10-10-x-switchHands-on tool on Test any new model on your own tasks: score a challenger against your current model with the switch rule.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#log-2026-10-10-x-caphttps://hayzlabs.com/reviews/changes/#log-2026-10-10-x-capHands-on tool on Your first API call: work out how many calls a spending cap buys.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#log-2026-10-10-merge-red-teamhttps://hayzlabs.com/reviews/changes/#log-2026-10-10-merge-red-teamRed-team your workflow is now a section of Checking agent work. The old address redirects there.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#log-2026-10-10https://hayzlabs.com/reviews/changes/#log-2026-10-10The site is live. If you are new to AI, begin with Start here.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#log-2026-10-10-merge-toolshttps://hayzlabs.com/reviews/changes/#log-2026-10-10-merge-toolsTools, connectors and structured output is now a section of Letting an agent near your email, browser and files. The old address redirects there.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#log-2026-10-10-rename-su01https://hayzlabs.com/reviews/changes/#log-2026-10-10-rename-su01What the lab runs on is replaced by Do I need to pay for AI, and for which plan?. The old address redirects there.Sat, 10 Oct 2026 12:00:00 GMThttps://hayzlabs.com/reviews/changes/#log-2026-10-10-merge-used-on-youhttps://hayzlabs.com/reviews/changes/#log-2026-10-10-merge-used-on-youWhen AI is used on you is now a section of Health, money and legal questions. The old address redirects there.Sat, 10 Oct 2026 12:00:00 GMT