OpenAI props Enterprise offerings with Private Link (security) and Projects (control)

Earlier this month, we discussed how OpenAI has some good numbers when it comes to Enterprise customers. Well, OpenAI is not looking to disappoint them as it has announced new updates to improve its services for enterprise customers. These features include better security, administrative control, and cost management for organizations building and deploying AI solutions.

Let’s start with security. OpenAI has worked upon data security and has introduced Private Link, a feature that can help in communication between a customer’s infrastructure and OpenAI’s systems. 

Why is it important? Because it can minimize exposure to the open internet. Multi-factor authentication (MFA) is now available as well. 

Both these features add on to the existing security protocols which includes SOC 2 compliance, single sign-on (SSO), data encryption, and role-based access controls. 

Now to the administrative control. OpenAI has introduced Projects. It gives organizations control over their AI initiatives within OpenAI’s platform, means that OpenAI is providing businesses with more detailed and specific ways to manage how they use AI tools and services offered by OpenAI.

Project owners and/or admins can assign specific roles and API keys, manage access to available AI models, and set usage and rate limits to prevent exceeding budgets. 

The Assistants API for developers also has new upgrades that has more file search capabilities, better file management, and support fine-tuned GPT-3.5 Turbo models.

OpenAI in now offering new options to help organizations manage the cost of their AI usage. It works throughput pricing provides discounts of 10-50% on sustained usage of GPT-4 or GPT-4 Turbo models. 

For non-urgent tasks, a new Batch API offers a 50% discount on asynchronous workloads.

More here.