# 7 Things SaaS Founders MUST Prepare for in the Age of AI Channel: Rob Walling Video: https://www.youtube.com/watch?v=n-XsTJv5kXs Duration: 7 min Language: English Words: 1364 Transcript page: https://viewrankai.com/tools/youtube-transcript/n-XsTJv5kXs --- [0:00] artificial intelligence open AI chat GPT Dolly how are we going to stay ahead of AI how are we going to build it into our SAS companies it's in the news every day and it's something a lot of SAS Founders are thinking about and it's something I've been doing a lot of thinking and talking about with SAS Founders something I've noticed is there are a lot of pitfalls a lot of ways to go wrong when using Ai and so in this video I'm going to talk about seven things SAS Founders should prepare for in this age of AI I'm Rob Walling I've started six companies five of them bootstrapped I've [0:32] invested in more than 150 startups and I've written four books about entrepreneurship something that startup Founders are going to have to be aware of as we start building AI into our products or the data and privacy concerns and there are a lot of different ways to think about this one is are you using an API like open AI to handle your AI processing and to get you the results in that case are your customers going to be okay with you taking whatever it is they enter into your application and passing it to open AI or whatever other provider that you use and then the other side of the coin [1:06] is if you decide to build your own models and you need training data are there issues with scraping data from the internet publicly available data as openai says there's a lot to uncover here and I believe there are already at least one court case probably several that are involving training data and data and privacy concerns around AI it's something that's going to continue unfolding and as startup Founders we have to be aware of that as we build AI into our applications a related point is point number two are we going to end up with gdpr style opt out for any application using Ai and if so is your [1:40] basic AI feature isn't actually going to add much to the application if folks can opt out of it and in that case is it better to not build it in the first place going to be a lot of questions in this video that are unanswered but these are things that as a SAS founder you have to weigh what's the effort the level of effort to integrate AI into your application and the level of value will it get more people to sell sign up and will it reduce churn and if the answer to those is definitely then obviously this may be something you want to take a risk on but keeping in mind [2:07] that gdpr type concerns and privacy and data privacy and those kinds of things are going to potentially be a bumpy road ahead Point number three to be thinking about is our AI hallucinations or potentially bad outputs from a feature you build into your application will they potentially erode customer trust if they don't understand that AI isn't perfect and do you need to communicate that in your interface that look sometimes AI is going to do kooky things it might hallucinate facts as developers and Tech folks who are using chat GPT on a daily or weekly basis I think some of us are you're likely to take what chat [2:43] GPT says with a grain of salt but lay people who might be using your application if you're focused on Realtors or you have Physicians or you have lawyers they may not understand that AI isn't perfect and it could potentially damage the relationship and the reputation of your software Point number four to keep in mind is is there potential risk in becoming too dependent on AI specifically I'm thinking with your internal processes what if you get to the point where you drop your Staffing way down on your customer support team because they're augmented by Ai and then what if there's a bug in that AI system or what if that system [3:18] goes down what if you aren't able to use it for a day or a week how do you cope with that in the short term realistically even with AI it's vital to maintain human oversight and some control think about if I was using AI to do customer service responses I could see a world where I get complacent and I don't even fully review the emails that AI is generating and it could send inaccurate data going out to customers if AI is right most of the time but you have to read 98 emails to see two that go out that just have crazy responses as humans aren't we gonna accidentally let [3:54] a few of those get through so most of the points so far have been questions or things to be concerned about about the the negative side of AI the dark mirror of AI if you will but 0.5 is does AI become table Stakes an expected feature in all SAS apps so as a Founder do you want to start getting on the wave early I think if you are not paying attention to how you can improve your internal operations with AI and how you could potentially integrate it into your product as customer facing features I personally think you're making a mistake because building AI into your product I [4:25] don't think it's just a buzzword you know this is not something like crypto and blockchain where some people are trying to force that into their SAS applications but most apps I think can use AI in some form or fashion even if it's not so obvious to the customer that it is AI they can still use it in a way to help their customers be awesome faster Point number six is to think about with the pace of change and how quickly AI is changing it seems like a normal pace of adoption would be years and this is happening in weeks or months and the normal pace of change is 5 or 10 [4:58] times faster with that in mind mind when I think about a piece of software that I'm going to build today that's going to run for a decade I think about features needing to be stable if openai has changed so much from 3.0 to 3.5 to 4.0 in its behavior and we could say it's getting better but really it's just becoming different so what happens in 4550 6070 does this break things that you've built into your app that you feel are stable but as these language learning models adapt and get better is there a potential for them to quickly become crafty or quickly just break things in your app overnight because any [5:33] Reliance on a third-party API always has the possibility to do that but most apis don't change this quickly in fact no apis change this quickly the seventh point to be aware of as a SAS founder is that AI really has a black box problem there's a lack of transparency in two ways first of all it can be challenging to understand how an AI system made a particular decision or a prediction so this lack of transparency can make it difficult to troubleshoot problem problems right or explain decisions to customers or prove compliance with regulations so that lack of transparency on that side is going to become an issue [6:08] as we move forward and the second side is when do you as a company need to indicate to a customer that something is AI versus human driven or AI versus code driven we don't know yet but these are things that are going to be playing out in the coming months and years if you've enjoyed this video you won't want to miss this next video about AI make sure you're subscribed to the channel like the video If you enjoyed it I'm putting out a new video every week about building launching growing SAS products and all things SAS Founders should be considering thanks for watching --- About this transcript Read from YouTube's own caption track and laid out by ViewRank AI (https://viewrankai.com). ViewRank AI finds the videos already beating a creator's own average on Instagram, TikTok and YouTube Shorts, transcribes them from the audio itself in more than 60 languages, and turns what worked into new ideas and scripts. Free transcript tools, no account needed: https://viewrankai.com/tools How to read any video this way: https://viewrankai.com/llms.txt