@AI Advantage Team Thanks Jaycebel, and I appreciate the thoughtful reply! 😊 But I think there's a bit of a contradiction hiding in there that's worth pulling apart. You say the goal isn't dependency, it's freeing up time for higher-value thinking. But "freeing up time" isn't a once-off transaction, in my view it's ongoing. Every day you lean on AI to do the groundwork, your workflow reshapes itself around AI being there tomorrow too. That IS dependency, whether we frame it nicely or not. A tool you occasionally reach for is different from a tool your whole operating rhythm has quietly rebuilt itself around. And on "amplifies capability rather than replaces it" I'd push back on the word "capability" itself (being an Business Architect I have to lol). Capability (output) and ability (the underlying skill) aren't the same thing, and a tool can grow one while eroding the other. GPS is the classic example: people lean on it constantly, their output (getting places) goes up, but their actual navigation ability measurably atrophies. You only find out how much ability you lost the day the tool isn't there. Which is exactly my point about Telstra. Nobody plans for "our billing system has a bad day" ,it's boring, not a Carrington event, but it still takes the tool away for a few hours. If a business has fully integrated AI, the honest question isn't "what if the sun does something dramatic," it's "what does Tuesday look like when the API is down for six hours." Most businesses don't have an answer to that yet, and I think that's worth sitting with rather than smoothing over. 💯

@Roy Shirley Best way to do it.

There is this saying they had "Paralisis by analysis", and you doing it right @Roy Shirley. Build and improve, instead waiting to be 100% polished. If you then take into consideration to ensure that ALL your functionality is build modulaire, so that functions can be taken out and inserted without breaking things. This in the industry (in my humble opinion) is often done wrong. You can have your app as multiple modules that all talk to a "bus". Eg System Integrated Infrastructure. Think of it all small apps talking to each other that make up you final app. This in turn not only has implementation, testing and development advantages but also authentication. Eg you can give access to each "module" based upoin users login (subscription). Anyways.... I am rabling now :-).

This is one of my "biggest" issues really. I do not have one... I only communicate online and via messaging (due to some brain issues I have). Verbal conversations are ... well ... not my thing anymore unfortunatelly. With that in mind, videos etc are a no go for me hence client acuisition is only possible for me to do naturally... bit like fishing.

@Jaskaran Saini Writing posts really and basically about climate and weather... mostly science style reports

@Ellen Monique Bosboom I asked chatGPT and as to "chatgpt cannot work with a regular link, so I need to find out what else to do ." if that was true. This is what it said. It depends on the type of link. Yes, I can work with a regular link if: - It's a publicly accessible web page. - It's a public YouTube video (I can often use available captions or information). - It's a publicly available PDF, document, or article. - The website doesn't require you to log in. I cannot directly access: - Videos stored on your computer via a local path. - Files behind a login (Google Drive, Dropbox, OneDrive, SharePoint, etc.) unless they are publicly shared. - Private YouTube or Vimeo videos. - Streaming services like Netflix, Stan, or Disney+. - Files that require authentication or special permissions. For video transcription If your goal is to transcribe a video, the most reliable options are: 1. Upload the video directly here. This gives the best results. 2. If the video is too large, split it into smaller parts (for example, 30–60 minutes each) and upload them. 3. If it's on YouTube and has public captions, send me the link and I may be able to use the available transcript. 4. If it's in Google Drive or Dropbox, download it to your computer first, then upload the video file here. If you tell me: - where the video is (YouTube, Google Drive, Dropbox, your computer, etc.), and - approximately how long it is, I can recommend the easiest way to get it transcribed.

@Ellen Monique Bosboom transcribing a video is not something you need AI for. Those who do a bit of coding use python whisper to transcribe the videos. If you confident enough you can get instructions from chatGPT on how to install it on your desktop/laptop. If you have many videos that is what I would do. Use "if I want to transcribe videos on my desktop not using ai what can I do can I use python?" should give you some guidance. This is what I have installed on my secured video network that replaces zoom (yes it is private). Users can record their video, some meetings go for hrs so the files get quite big. 1. A python script transcribes the meeting. 2. I feed the transcribe file into AI (running local for security reasons). 3. AI generates meeting minutes, with action items, summary cards etc. 4. User (moderator only) can download the meeting recording, transciption, meeting minutes. So basically if you want to transcribe a lot of videos then doing locally with python (I use python there might be others) is likely the best way forward.

Some businesses are actually hiring the staff they replaced with AI back...I guess they realised garbage in garbage out. We now see AI learning from AI, which means that articles and items written by AI and published on the internet (eg by you or me) are used by AI to learn.... (it is a quantity game really, there is no real quality assurance on AI learning).

@Harsh Singh Very much so, in relation to how the a response is formulated. Yet the content of the response is based upoin what it learned and trained upon (as you know). If you train ai internal, then yes you have some influence on what it knows, however there will likely a large part of users that will use the "run of the mill" type AI like chatGPT, Claude etc. these are trained (from what I understand) on internet available data. This is where the garbage in and garbage out comes in.