April 13, 2023

Trust

How Healthcare Providers Can Increase Patients’ Trust in Artificial Intelligence

Entrepreneur, 04/13/23. AI has plenty of promise for healthcare but there are concerns about privacy, security, and bias. Patients prefer AI for administrative tasks over personal tasks.

To increase trust in AI, providers should be transparent and ethical, guarantee data privacy, and educate patients. For the public, factors such as representation, user reviews, ease of understanding, trialability, communication, and socialization and bonding build trust. READ MORE

Forecast

Michael Dell On How AI Is Reimagining Business As ‘The Cost of Cognitive Power Goes To Zero’

CRN, 04/13/23. The fastest-growing workloads in Dell’s ISG unit are AI- and machine intelligence-driven. Dell believes that AI can improve his company’s business by handling customer problems and employee questions more effectively. Solution providers such as Simnet and Pia are already using AI to improve service quality and increase customer satisfaction. Dell has introduced PowerEdge servers that support AI model processing. READ MORE

fashion

3 ways fashion ecommerce will evolve through generative AI and ChatGPT

INSIDER INTELLIGENCE, 04/13/23. Generative AI can transform the fashion industry by providing personalized experiences for customers. It can solve challenges related to buying clothes online, and brands can use it for product search and discovery, description, and customization.

Transparency and quality control are crucial as is addressing intellectual property issues. McKinsey estimates it could add up to $275 billion to the fashion industry’s profits. READ MORE

Limitations

AI chatbots don’t actually “know” anything. Can we fix that?

Freethink, 04/13/23. ChatGPT’s success has led to an AI arms race, but large language models (LLMs) have limitations. To address these, research groups are working on augmented language models (ALMs), such as chain-of-thought and retrieval-based models.

While ALMs hold promise for enhancing productivity, the issues with LLMs aren’t expected to be entirely solved soon. READ MORE

Search

The AI bot has picked an answer for you. Here’s how often it’s bad.

The Washington Post, 04/13/23. AI chatbots like ChatGPT, Bing, and Google’s Bard are the future of online searching. The problem is they deliver definitive answers even when they’re confidently wrong.

Bing’s AI bot has questionable research practices, such as choosing fringe sources and not understanding what a source has to say. One potential solution is to tune the AI to more often say “I don’t know.”

Companies should be more honest about their limitations and change the design to make this clearer. READ MORE