I thought I'd share this because it might be helpful to some of you, especially if you care about privacy. So as you probably know one of the drawbacks of using LLMs, web based and API, is that almost all major ones, like OpenAi/Anthropic/Gemini etc. retain your input and output data for 30 days. In business and even personal uses it's a problem because you don't have data privacy plus there's additional risk of your data being stolen from their servers. So you have an option... to run a local llm, like Google's Gemma/Nvidia Nemotron/OpenAI GPT OSS/Qwen/Deepseek etc. and everything stay on your computer. I actually tried it few days ago :) but quickly found out that you need a powerful machine to run it efficiently. Think a $15-20k workstation not my laptop lol. It worked but a query that I do on ChatGPT for example, which took 1 minute to complete, took 65 minutes on my laptop 😄. In addition local llms are not nearly as good as API models. But they are good enough for many simpler tasks where top notch advanced reasoning is not needed. And if you/your client can invest in a $15/20k machine to run it it's an option. So I dug deeper and found a more affordable option :). Perplexity offers Sonar/Sonar Pro API. It has Zero Data Retention Policy. None of your inputs/outputs are retained, only metadata. https://docs.perplexity.ai/docs/resources/privacy-security Hope this helps!

Hey guys! Just a quick reminder for paid members that the AI Audit Masterclass with Corey Ganim is happening TODAY at 1:30 PM US Central Time. This will also be recorded. You can find the link to join on the Calendar - https://www.skool.com/ai-marketing-insiders/calendar Corey will go in the weeds and any questions are fair game to pick his brain on how to make this 'AI Audit' business model successful. This will be super valuable. Again these bonus Masterclasses are for Paid Members ONLY. This means both VIP and Premium members. Look forward to seeing everyone!

I'm sure we can all relate to this, especially in marketing! 🤣 Some light Sat humour. https://www.youtube.com/watch?v=T4Upf_B9RLQ [Edit] - Here's the link to what the video is all about https://www.forbrukerradet.no/breakingfree

Of course right after I get off our weekly community call for VIP Members I see this lol - https://openai.com/index/previewing-gpt-5-6-sol/ Likely a direct competitor to Claude Mythos & Fable models. Let's see if the US Government bans them lol. But unfortunately in terms of access, this is from OpenAI - 'During the preview, GPT‑5.6 models will initially be available through the API and Codex to a select group of trusted partners and organizations. We plan to make them more broadly available to people using ChatGPT, Codex, and the API soon.'

We're halfway through 2026 already. Back in January I asked what was actually working for people three weeks in (that post is here: https://www.skool.com/ai-marketing-insiders/3-weeks-into-january-whats-actually-working?p=73f711f7). Six months and a LOT of updates later (Claude alone has shipped more changes than I can keep track of, never mind everything else), it feels worth asking again. So, for the newer members especially: what's genuinely changed for you this year? Not what you've tried, but what you're actually using. What's improved, what's clicked, what's a workflow or idea the rest of us might find useful? Here's what's shifted for me in six months: - The efficiency gains haven't slowed down. If anything they've picked up. - Ignore the posts telling you that if you only 'chat' with AI you're doing it wrong. It completely depends on what you're chatting about. Using Claude Code as a thought partner is still one of my biggest gains. I can take something messy, get right into the weeds on it, and come out the other side with real clarity. - Talking things through with friends and colleagues is still good, and I still do it. But for ideas, detail, structure and clear communication, AI is hard to beat. We're often vague when we're put on the spot ("let me have a think", "I'm not sure", "maybe we could..."), and then you wait days for a reply. With AI there's no waiting, and you still get the back and forth. You can push back, disagree, ignore half of what it says, and still come away with gems a person would rarely hand you on the spot. - Skills have matured. I can capture a workflow once and trigger it with a single short phrase, instead of re-explaining it every time. That's a big one for me, because how I describe things varies day to day, and a skill takes that variability out. - The built-in memory and bigger context windows have caught up with the external memory setup I built back in January. It now holds the thread of a whole project between sessions, so I'm re-briefing it far less. - MCP connections and integrations have come a long way. More and more I can stay inside Claude Code instead of hopping between tabs and tools. That matters more than it sounds. Constant context switching is genuinely draining, whatever the productivity crowd says, and fewer browser tabs is a real win for me. - I can also set several agents going at once on separate jobs now, which has saved real time on the bittier work. - The loops idea I posted about recently (here: https://www.skool.com/ai-marketing-insiders/boris-cherny-doesnt-prompt-claude-anymore-he-runs-loops-so-whats-a-loop?p=b494666b) is interesting, and the steady stream of updates like that has been invaluable, not because I adopt every one, but because they help me work out where each thing actually fits. Loops might only suit a small slice of what I do, and that's fine. You don't have to take on everything, or take it on fully.