Generative AI Offerings

Generative AI is revolutionizing healthcare at UTHealth Houston by streamlining workflows and enhancing decision-making processes. By automating routine tasks, it allows healthcare professionals to focus more on patient care and innovative research. Additionally, generative AI aids in the rapid analysis of vast datasets, leading to more accurate diagnoses and personalized treatment plans. This cutting-edge technology is driving UTHealth Houston towards a future of more efficient and effective healthcare delivery. Below is a table featuring all the current offerings at UTHealth Houston as well as their benefits and related costs: 

Click Icon to redirect to app page.

	Microsoft CoPilot (Free Version)	Microsoft CoPilot	Azure AI Studio	OpenAI API Access	ChatGPT4o
Click Icon to Redirect to Application Website
Price	Free	$30mo. / User Billed Annually Request a License	Variable Pricing Request a License	Per Token Charge Request a License	$30mo. / User Billed Annually Request a License
Personal, Confidential, Private and Proprietary Information	No	Yes with License	Yes after proper configuration	Yes for zero retention eligible endpoints (More Info)	Yes with License
FERPA / HIPPA Protected Information	No	Yes with License	Yes after proper configuration	Yes for zero retention eligible endpoints (More Info)	Yes with License
Features	Basic Document Assistance with grammar, spelling and formatting	Use an end-to-end conversational design studio to extend and customize Copilot for Microsoft 365 or build your own copilot	Build your own tools and application with customized models	Direct access to OpenAI’s foundation models to create custom applications	Ability to create custom GPTs
	Limited integration with Microsoft Word, Excel and Outlook	CoPilot combines a specialized language model with security-specific capabilities to incorporate a growing set of security-specific skills informed by unique global threat intelligence	Built-in security controls and unique threat intelligence helps identify and protect against rapidly evolving threats.	Accepts as input any combination of text, audio, image, and video and generates any combination of text, audio, and image outputs.	Fast responses and analysis of large data sets.  GPT-4o is 9 times faster than GPT-3.5, with an average response latency of 0.32 seconds.
		Integration with Microsoft 365 Suite	Access to multiple foundational models (OpenAI, Anthropic, Meta’s Llama)	With GPT-4o, we trained a single new model end-to-end across text, vision, and audio, meaning that all inputs and outputs are processed by the same neural network	GPT-4o shows a 40% reduction in generating nonsensical or inaccurate information compared to GPT-3.5.
		Image generation, document summaries, e-mail generation	Ability to create custom retrieval augmented generation (i.e. chatbots) based on your own datasets	GPT-4o achieves GPT-4 Turbo-level performance on text, reasoning, and coding intelligence, while setting new high watermarks on multilingual, audio, and vision capabilities.	Improves conciseness by better understanding and processing context, logic, and causality, resulting in more accurate and insightful responses.