Congrats on the launch, spreadsheets are getting lots of AI upgrades these days, exciting!
If i were to try this out with some somewhat sensitive company data, what is the security profile of this? Would it potentially leak the data to MCP servers? Do I have control?
Hi Nicho, Andy (CTO Sourcetable) we have gone to extreme lengths to protect your security. We use a form of escrow encryption in maximum security mode to only send data through at the time of you using it. Data that you have selected in your spreadsheet or called upon in the context of the conversation can be used for mcp conversations. If you are really worried about sensitive data, you can always create a new sheet and only include data that you are 100% sure of. MCP is generally outbound (you choose what to ask) so unless you've directed the AI to do something that the MCP endpoint requires you should be ok.
Let me get this straight. AI, one of whose weaknesses is a tendency to randomly come up with creatively buggy, yet syntactically correct code, is meant to be editing spreadsheets, which are the least debuggable kind of computer program imaginable, one where even very serious bugs often have no visually obvious effect, one where the only to actually see the code is in tiny snippets, by clicking through every cell.
Congratulations on the launch. We’re spending on paid ads and optimizing ads is a pain. Fo you think sourcetable can help us make sense of optimization data and run experiments? I want to enable our analysts to take faster decisions.
Separate to the Superagents launch here, LLMs are excellent for keyword optimizations since the compression/summary/synthesis essentially comes for free out of the box. This isn't unique to Sourcetable, but I do find it extremely pleasant that vector analysis with LLMs is easy, not hard. SEM/SEO is all just math at the end of the day.
The main things we bring to the table are that the AI can write code and handle much larger datasets than fit in ChatGPT, etc., and also that Superagents you can pipe your data in without code or SaaS interface kludge, so you can ask much more complicated questions than you usually might if you're not great at cleaning, filtering or analyzing data.
To answer skyzouwdev - not sure why HN removed your post - we can connect to anything on the Internet. Popular services with better documentation tend to perform better. If you get stuck, you can augment and assist our AI by pasting instructions in the chat.
If i were to try this out with some somewhat sensitive company data, what is the security profile of this? Would it potentially leak the data to MCP servers? Do I have control?
The main things we bring to the table are that the AI can write code and handle much larger datasets than fit in ChatGPT, etc., and also that Superagents you can pipe your data in without code or SaaS interface kludge, so you can ask much more complicated questions than you usually might if you're not great at cleaning, filtering or analyzing data.