Mike Caulfield has released what he calls a Deep Background GPT. It is an AI fact-checking tool, available to all as a completely free GPT. Mike says:
I just released a (largely) non-hallucinating rigorous AI-based fact-checker that anyone can use for free. And I don’t say that lightly: I literally co-wrote the book on using the internet to verify things. All you do is log into ChatGPT, click the link below, and put in a sentence or paragraph for it to fact check.
The Anapoly team met this morning to talk about progress, harmonise our thinking, and firm up the plan for the next few weeks. It brought home the value of face to face discussion, because two interesting new ideas popped out of some animated debate.
When discussion crystallised something worth recording, we paused for a few moments to dictate a note into my mobile phone. The app transcribed it in real time, but with poor punctuation and various transcription errors (I need to get a better app, perhaps!).
Later, I gave the transcript to ChatGPT with the instruction: “The file 2025-06-28 meeting notes contains dictated notes from a meeting earlier today. Please give them a sensible format and correct obvious faults. If in doubt about anything, ask me to clarify.”
ChatGPT did not need clarification. In the blink of an eye it produced a set of well written, clearly organised notes that would have taken me twenty or thirty minutes to produce to the same standard.
OpenAI has been quietly developing collaboration features for ChatGPT that would let multiple users work together on documents and chat about projects, a direct assault on Microsoft’s core productivity business. The designs have been in development for nearly a year, with OpenAI’s Canvas feature serving as a first step toward full document collaboration tools.
This would be a change towards something like simultaneous collaboration with colleagues on a shared document in Microsoft Office. At present, a ChatGPT Team account allows more than one person in the Team account to work in a project space and to take part in the chats within that project, but only one person at a time, as I understand it.
A few hours spent with ChatGPT-o3 resulting in good first draft of a framework for thinking about our labs. It covers
types of lab
the roles of people involved with the labs
the core technical configuration of a lab
assets needed to launch, operate, and archive a lab
a naming convention for these assets
No doubt the framework will need to be tweaked and added to as our ideas mature.
The chat with o3 was a valuable mind-clearing exercise for me, and I was impressed by how much more “intellectual” it is compared to the 4o model. Like many intellectuals, it also displayed a lack of common sense on occasions, especially when I asked for simple formatting corrections to the canvas we were editing together. The 4o model is much more agile in that respect.
During the chat, when the flow with ChatGPT didn’t feel right, I hopped back and forth to consult with Perplexity and NotebookLM. Their outputs provided interestingly and usefully different perspectives that helped to clear the logjam.
A decision arising from my joint AI consultation process was the choice of Google Workspace for the office productivity suite within our labs. This will allow for much better collaboration when using office tools with personal licences than would be the case with Microsoft Office 365. Given the ad hoc nature of labs and the cost constaints we have, this is an important consideration.
… what I can say is that a theme throughout this self-analysis is this: I find ChatGPT to be a really useful tool when I already have some idea of what I want to do and when I’m actually engaged with the issue. I find it much less reliable or useful for completely automating parts of the process. There’s no substitute for reading the paper.
To identify and configure an AI toolkit for Anapoly AI Labs that credibly models the use of general-purpose AI tools in a small consultancy setting.
Author and date
Alec Fearon, 24 June 2025
Participants
Alec Fearon, with Ray Holland and Dennis Silverwood in email consultation ChatGPT-4o
Lab configuration and setup
This setup models a real-world micro-consultancy with three collaborators. It assumes limited budget, modest technical support, and a practical orientation. We aim to reflect the toolkit choices we might recommend to participants in Anapoly AI Labs sessions.
Preamble
If Anapoly AI Labs is to be a credible venture, we believe it must model the behaviour it explores. That means our own internal work should demonstrate how small teams or sole traders might use AI tools in everyday tasks – writing, research, analysis, and communication – not just talk about it. This lab note outlines our proposed working configuration.
Procedure
We identified common functions we ourselves perform (and expect others will want to model), for example:
Writing, summarising, and critiquing text
Researching topics and checking facts
Extracting and organising information from documents
Sharing and collaborating on files
Managing project knowledge
We then selected tools that:
Are available off the shelf
Require no specialist training
Are affordable on a small-business budget
Can be configured and used transparently
Findings
Core Tools Selected
Function
Tool
Licence
Notes
Writing & prompting
ChatGPT Team
£25–30/m/user
Main workspace for drafting, reasoning, editing
Search & fact-checking
Perplexity Pro
$20/m/user
Fast, source-aware, good for validating facts
Document interrogation
NotebookLM
Free (for now)
Project libraries, good with PDFs and notes
Office apps
MS 365 or Google
£5–15/m/user
Matches common small business setups
Visual inputs
ChatGPT Vision
Included with ChatGPT
Used for images, scans, and screenshots
Discussion of findings
This configuration balances affordability, realism, and capability. We expect participants in Anapoly AI Labs to have similar access to these tools, or to be able to get it. By using these tools ourselves in Anapoly’s day-to-day running, we:
Gain first hand experience to share
Create reusable examples from real work
Expose gaps, workarounds, and lessons worth documenting
We considered whether personal licences could be shared during lab sessions. Technically, they can’t: individual ChatGPT and Perplexity licences are for single-user use. While enforcement is unlikely, we’ve chosen to adopt the position that participants should bring their own their own AI tools – free or paid – to lab sessions as part of the learning experience. This avoids ambiguity about licencing and sets the ethical standard we want to maintain.
Conclusions
This toolkit would enable us to model our own small-business operations, treating Anapoly itself as one of the lab setups. That would reinforce our stance: we don’t claim to be AI experts; we’re practitioners asking the questions small businesses wish they had time to ask, and showing what happens when you do.
Recommendations
Configure project workspaces in ChatGPT Team to reflect different lab contexts
Maintain prompt libraries and reasoning trails
Make costs, configurations, and limitations explicit in diary and lab notes
Evaluate whether to add AI-enhanced spreadsheet or knowledge tools (e.g. Notion, Obsidian) in future iterations
Tags
ai tools, toolkit, configuration, modelling, small business, chatgpt, perplexity, notebooklm, office software, credibility
Glossary
ChatGPT Team – OpenAI’s paid workspace version of ChatGPT, allowing collaboration, custom GPTs, and project folders. NotebookLM – A Google tool for working with uploaded documents using AI, currently free. Perplexity Pro – A subscription AI assistant known for showing sources. Vision input – The ability to upload images (photos, scans) and have the AI interpret them.
The EU AI Act establishes a risk-based classification system for AI systems. Compliance requirements depend on the risk the system poses to users. Risk levels are set at unacceptable and high. General-Purpose AI systems like ChatGPT are not classified as high-risk but are subject to specific transparency requirements and must comply with EU copyright law.
Also, the Act “aims to support AI innovation and start-ups in Europe, allowing companies to develop and test general-purpose AI models before public release. That is why it requires that national authorities provide companies with a testing environment for AI that simulates conditions close to the real world. This will help small and medium-sized enterprises (SMEs) compete in the growing EU artificial intelligence market.“
Unlike the EU, the UK has no general-purpose AI sandbox. The UK takes a sector-led approach to AI oversight, relying on existing regulators to operate their own sandbox initiatives under the government’s pro-innovation framework. Each sandbox is designed around the compliance needs and risk profiles of its domain. Existing sandboxes are focused on sector-specific or compliance-heavy contexts, for example:
FCA AI Sandbox (2025) – Financial services only; supports firms developing or integrating AI tools into fintech workflows.
ICO Regulatory Sandbox – Suitable for testing AI applications involving personal data, especially where GDPR-like safeguards are needed.
MHRA AI Airlock – For AI used in medical devices.
These UK sandboxes are geared toward testing purpose-built AI tools or integrations into regulated industry IT systems. There is no current sandbox designed for SMEs exploring general-purpose AI like ChatGPT in everyday, low-risk tasks. While tools like ChatGPT could be included in sandbox trials (e.g. via the ICO for data privacy concerns or the FCA for financial services), these environments are not designed for routine or everyday use. They are structured for defined compliance challenges, not routine experimentation, and they require a defined project with specific compliance aims.
A search by ChatGPT found no evidence of any current provision for open-ended, exploratory use of general-purpose AI by SMEs in unregulated or lightly regulated workflows. Anapoly AI Labs can occupy that space, modelling practical, credible AI use outside formal regulation.
In that context, Anapoly’s modelling approach could include a cycle of experimentation followed by proof of concept (POC). A proof of concept is a small-scale test to check whether an idea works before investing in full implementation.
Transparency Label Justification. This piece was authored by Alec Fearon. ChatGPT was used to assist with regulatory research (e.g. the EU AI Act and UK sandbox landscape), structure the argument, and refine language. The ideas, framing, and conclusions are Alec’s. AI contributed supporting detail and wording options, but did not generate or structure the content independently.
It documents an internal experiment to refine the custom instructions we use with ChatGPT – what we expect from it, how it should respond, and how to keep it useful across different tasks. The aim is to define a persona for the AI assistant that is more consistent and can adapt to the different types of assistance required of it.
It’s a good example of how we’re using the Labs: not to explain AI, but to find out what it’s actually good for.