NVIDIA Open Sources NeMo Guardrails
Written by Alex Denham   
Thursday, 01 June 2023

NVIDIA has released open-source software aimed at helping enterprises keep applications built on large language models aligned with their safety and security requirements.

NeMo Guardrails can be used to monitor smart applications powered by large language models (LLMs) to make sure their output is accurate, appropriate, on topic and secure.

nvidialogo

The software includes all the code, examples and documentation businesses need to add safety to AI apps that generate text.

NVIDIA designed NeMo Guardrails to work with all LLMs, such as OpenAI's ChatGPT. Organizations are finding that the use of AI sometimes results in the output of chat software being inappropriate, as the models can become sexist, racist or take extreme political views, for example. NeMo lets developers align LLM-powered apps so they're safe and stay within the domains of a company's expertise.

The toolkit can be used to set up three kinds of boundaries - topical guardrails to prevent apps from veering off into undesired areas; safety guardrails to help ensure accurate and appropriate responses; and security guardrails to control what apps connect to.  NVIDIA says the topical guardrails might be used to prevent customer service assistants from answering questions about the weather, while the safety guardrails can filter out unwanted language and enforce that references are made only to credible sources. Finally, the security guardrails restrict apps to making connections only to external third-party applications known to be safe.

Since NeMo Guardrails is open source, it can work with all the tools that enterprise app developers use.  NVIDIA says that for example it can run on top of LangChain, an open-source toolkit that developers are using to plug third-party applications into the power of LLMs. NeMo Guardrails is also designed to be able to work with a broad range of LLM-enabled applications, such as Zapier.

NVIDIA is incorporating NeMo Guardrails into the NVIDIA NeMo framework, which can be used to train and tune language models using a company's proprietary data. Much of the NeMo framework is already available as open source code on GitHub. Enterprises also can get it as a complete and supported package, part of the NVIDIA AI Enterprise software platform.

NeMo is also available as a service as part of NVIDIA AI Foundations, a family of cloud services for businesses that want to create and run custom generative AI models based on their own datasets and domain knowledge.

nvidialogo

More Information

NVIDIA NeMo

Related Articles

NVIDIA Broadcast 1.4 Adds Eye Contact  

Nvidia Releases Updated AI Framework

Nvidia Releases Updated AI Framework

NVIDIA's Turing - A Big Leap Forward For GPUs

NVIDA Updates Free Deep Learning Software

To be informed about new articles on I Programmer, sign up for our weekly newsletter, subscribe to the RSS feed and follow us on Twitter, Facebook or Linkedin.

Banner


GR00T Could Be The Robot You Have Always Wanted
27/03/2024

We may not have flying cars, but we could well soon have robots that match up to predictions for the 21st century. Nvidia has announced GR00T, a cleverly named project to build robots using foundation [ ... ]



JetBrains Announces TeamCity Pipelines
19/03/2024

JetBrains has released a public beta of TeamCity Pipelines, a cloud-based Continuous Integration/Continuous Deployment (CI/CD) service for small and medium-sized engineering teams.


More News

raspberry pi books

 

Comments




or email your comment to: comments@i-programmer.info

Last Updated ( Thursday, 01 June 2023 )