@sidraref congrats on the launch! seems like something setup as a compliance tool for companies that interact with LLMs, are you thinking of it that way? More thoughts

@frank_denbow, first of all, thank you for your review. I believe we need to do a better job of highlighting all our value propositions in the hero section. You are right; our product assists companies with their compliance needs. To address some of the questions you raised: 1) ZeroTrusted.ai acts as middleware between users and Large Language Models (LLMs) through our secure chat or API. 2) There's no need for you to have a separate account or key for each LLM. Instead, we provide our own keys, allowing access without revealing your identity to the LLMs. 3) Your point about the potential exposure of scrubbed data in the event of a breach at a third-party LLM's network is partially correct. However, your identity will not be linked to this data. This scenario is particularly critical for both individual users and businesses. 4) We offer features that maintain context when sanitizing sensitive data. Example #1 Suppose a medical expert needs to process the following input: "Create a summary of this patient's diagnosis: Patient Name: Paul Smith Date of Birth: 12/08/1987 SSN: 666-555-5555 Diagnosis: Chronic Heart Disease Treatment Plan: Undergo heart surgery by Dr. Smith at General Hospital on 03/15/2024 Insurance: ABC Health Insurance, Policy Number 434444444" Before we pass it on to any LLM, we'll alter the sensitive (PHI compliance violations) details and transform to: "Create a summary of this patient's diagnosis: Patient Name: John Doe Date of Birth: 01/01/1970 SSN: 123-45-6789 Diagnosis: Chronic Heart Disease Treatment Plan: Undergo heart surgery by Dr. Smith at General Hospital on 03/15/2024 Insurance: ABC Health Insurance, Policy Number 987654321" This way, we preserve the context, while ensuring that your PHI compliance is met and sensitive data isn't exposed to LLM. After getting a response from the LLM, we revert the swapped sensitive value to your original while preserving the result context. We will be adding features that will allow RLHF to learn and adjust to customer preferences. Example #2: if a customer (individual or corporate) submits legal content containing sensitive information, our "Fictionalize" feature replaces this data with fictional values. This ensures the preservation of context while protecting the actual sensitive information. The original data is restored once a response is received. Hope this helps. Please us know if there are any more questions you would like us to address. Once again we appreciate your feedback and we hope your views can gain from the privacy and security solution that we provide.

@xavier_jam thank you. well appreciated!

@10xmarketer Glad you liked it. Thanks.

@af1que1 thanks for the great feedback. We are constantly working hard to ensure the highest level of privacy and security.

@anthony_latona Currently, it's hard for us to show the history like ChatGPT since we don't store it anywhere. The history is only visible per session and it's securely stored in session on your browser, so it'll be cleared after the logout. So, it's the limitation that we've exposed for security reasons. That's the most limitation as of now. Thanks! Glad you liked it.

@erburgess we would love to talk and explore opportunity. Please message us at contact@zerotrusted.ai

@elisa_tamburini thank you very much

Hey @abdellah_abbous , Here are your answers, - We currently support ChatGPT 3.5, ChatGPT 4, Claude, Llama 2, CoHere & Palm LLMs. - We have Zapier integrations to automatically send chat reports. For example, if you have an automation to automatically create blog posts using ChatGPT and post it in your site, then your Zapier integration can be used in between to see if the ChatGPT generated post is Plagiarism & Copyright free as well as we give an AI text detection score to make your blog/posts stronger. - Additionally, we have our Adapter library using the LangChain which allows you to quickly integrate our APIs. I would love to show you a demo on Zapier integrations and LangChain integration, please reach out to contact@zerotrusted.ai to initiate a discussion on this.

@siddharth2395 Yes, Privacy and LLMs summarization are the main points. Thanks for the review!

Thank you @sandradjajic. We agree. It’s extremely crucial to feel a sense security while using technology.

@bonvisions We find out sensitive information in your prompt and then we sanitize it before passing it over to LLM. For example, your prompt: Hey ChatGPT, my name is Bon Visions, My credit card information is 123-4567-8910. Give me payment integration code for my site. Our modified prompt: Hey ChatGPT, my name is John Doe, My credit card information is 111-4444-9999. Give me payment integration code for my site. Now we pass the modified prompt to the ChatGPT, so your sensitive information is passed sanitized and thus protected.

@rukhsar_amjad Sure, give it a shot! We have a lifetime free subscription.

@sentry_co we support openAI and many other LLMs. In our premium plan, we enhance accuracy by running your query on multiple LLMs and then we use our scoring model to determine most reliable - think Kayak. Users can also select 1 LLM. Our main role is to be the privacy layer for our customers.

@femitfash If data winds up at OpenAI. Then privacy is broken. There is only so much you can do with obfuscation. Or am I missing something?

@sentry_co this is partly true, however: 1) any sensitive information are already masked by our system. 2) most importantly, your identity isn't tied to that query you searched and it cant be traced back to you since we don't keep any data and wont even know who sent what. Your queries are just so anonymous text embedded within large corpus amount of data. BTW - this is only aspect of the value we provide.

@femitfash Right. But human data is extremely identifiable. And if you obfuscate it too much, then OpenAI doesn't understand it right? or? Im just trying to understand here. Some kind of illustration or example would make this more clear I think. A segment of text goes into your service then goes to openai etc. What does the data look like at each juncture etc? You know what I mean? Btw. Love the problem your trying to solve here!

@sentry_co Here is an example - suppose a medical expert needs to process the following input: "Create a summary of this patient's diagnosis: Patient Name: Paul Smith Date of Birth: 12/08/1987 SSN: 666-555-5555 Diagnosis: Chronic Heart Disease Treatment Plan: Undergo heart surgery by Dr. Smith at General Hospital on 03/15/2024 Insurance: ABC Health Insurance, Policy Number 434444444" Before we pass it on to any LLM, we'll alter the sensitive (PHI compliance violations) details and transform to: "Create a summary of this patient's diagnosis: Patient Name: John Doe Date of Birth: 01/01/1970 SSN: 123-45-6789 Diagnosis: Chronic Heart Disease Treatment Plan: Undergo heart surgery by Dr. Smith at General Hospital on 03/15/2024 Insurance: ABC Health Insurance, Policy Number 987654321" This way, we preserve the context, while ensuring that your PHI compliance is met and sensitive data isn't exposed to LLM. After getting a response from the LLM, we revert the swapped sensitive value to your original while preserving the result context. We will be adding features that will allow RLHF to learn and adjust to customer preferences. Hope this helps.

@dayton_thorpe thanks for the awesome feedback. We absolutely agree :)

@johny_burg While integrating for any business needs, if you use the Zapier integration, it'll be a low-code/no-code solution. It'll be easy for you to quickly integrate and maintain.

@sophia_watt we utilize LLM ensemble - we essentially run your query through multiple LLMs and use our scoring model to rank the result. This is quite challenging but our results great and they reduce like hood of data poison when compared to direct output from any single LLM. We are constantly working to improve this process to become as accurate as possible.

@maali_1 Glad you liked it! Thank you so much.

@amit_arora Absolutely. Thanks! Did you give it a try? We have a lifetime free plan.