PYMNTS MonitorEdge May 2024

This Week in AI: Enterprise Intelligence, Healthcare Efficiencies, Deepfake Threats

LLM, Large Language Model, AI, Artificial Intelligence

Artificial intelligence (AI) is becoming the standard-bearer for innovations that progress in leaps and bounds.

Putting an exclamation point on the technology’s value to nearly all sectors and industries, on Tuesday (Feb. 13), chipmaker Nvidia exceeded Amazon in market value for the first time since 2002 as more companies look to grow their AI offerings and consumer interest in the technology increases.

And AI took over the Super Bowl this past Sunday (Feb. 11) with AI system providers and developers spending millions on prime-time ad spots to promote AI’s capabilities and showcase the innovation’s integration across a variety of use cases.

Crucially, for those still associating AI capabilities with text-generation and chatbot responses, ecosystem pioneer OpenAI on Thursday (Feb. 15) launched Sora, a new text-to-video model that can create videos up to one minute long based on text prompts from users.

This is the must-read pulse check on the top AI news and innovations PYMNTS has been tracking this week. 

AI Chatbots Are Coming for Business Processes

Building on the consumer-facing success of chatbot products, AI systems are now looking at the attractive economics of enterprise AI solutions.

In a move that send ripples throughout the AI landscape, OpenAI’s Chairman Bret Taylor on Wednesday (Feb. 14) launched Sierra, an AI chatbot startup, aiming to develop conversational AI agents specifically designed for businesses. He dismissed concerns about potential competition by claiming that Sierra operates at a different layer of the AI stack compared to OpenAI’s ChatGPT.

But Sierra isn’t the only AI chatbot for business out there. 

PayPal-backed Rasa also on Wednesday announced it had raised $30 million in a Series C round to grow its generative conversational AI platform for enterprises. The company aims to redefine how businesses use generative AI-powered chat and voice platforms at scale.

Nokia has also debuted a suite of AI-powered tools for industrial workers, letting them “understand complex machines, get real-time status information and industries to achieve greater flexibility, productivity, sustainability, as well as improve worker safety,” Nokia said.

Enterprise messaging platform Slack is also joining the party, debuting Slack AI, its AI tool to help users find information more easily and boost worker efficiency.

Still, gaining efficiencies from AI is perhaps easier said than done. Microsoft’s AI assistant, Copilot for Microsoft 365, has been under scrutiny by testers for over six months, with reviews reportedly a mix of positive and negative.

Building Better AI Systems Delivers Better Results 

We are still in the early innings of the AI era, and PYMNTS covered on Monday (Feb. 12) how the ecosystem’s pace of development is already shattering Moore’s Law. 

But all the news and product launches capturing headlines that doesn’t mean that genAI systems, in their current state, are perfect. That’s why Google and Anthropic are working to address the limitations of their genAI systems, including issues around hallucinations, copyright and sensitive data, which have raised concerns among potential customers. 

One week after its rebrand as “Gemini,” Google’s AI tool received a new update dubbed “Gemini 1.5.”

And on Tuesday (Feb. 13), OpenAI announced it is testing a ChatGPT product with enhanced user and interaction memory. 

Alphabet, the parent company of Google, experienced a decline in its shares up to 3.8% Thursday (Feb. 15) following a report that OpenAI, the owner of ChatGPT, is developing a web search product that would directly compete with Google.

And news broke Thursday (Feb. 15) that Apple is reportedly working on a new AI-powered coding tool for XCode, its flagship programming software.

Thanks to AI and digital platforms, Michael Boeke, vice president of payment product management enterprise at CCC Intelligent Solutions, told PYMNTS on Thursday that, “you’re getting to a point where you essentially have ‘no’ steps that are required to complete a transaction.” 

Controlling for the Damage AI Deepfakes Can Sow 

The Federal Trade Commission (FTC) on Thursday announced it is proposing a new set of rules that would prohibit the impersonation of individuals, explicitly noting AI’s impact on worsening the impact of fraud.

This comes as consumers lost a record $10 billion to fraud in 2023, PYMNTS reported last week, and it was reported Wednesday (Feb. 14) that scammers are turning to genAI to improve and scale their phishing attacks.

And the rise of AI-powered deepfake technology has fostered a corresponding boom in state-level legislation, with nearly half of every state legislature in session considering bills designed to ward off the threat of deepfake technology.

As PYMNTS covered during the lead up to Valentine’s Day, online relationships have become a breeding ground for exploitation, with fraudsters using AI to capitalize on the vulnerabilities of hopeful hearts for financial gain. 

But it’s not all bad news about bad actors. OpenAI, with its partner Microsoft, have disrupted state-sponsored hackers attempting to use its AI technology for malicious purposes.

Why AI and Healthcare Are a Perfect Pair 

PYMNTS on Wednesday also unpacked the intersectional opportunities between connected wearable device health data and its potential to fine-tune and personalize health LLMs (large language models).

The coverage builds on previous reporting on Monday about how AI is helping build smarter hospitals that deliver better patient care.

And it’s not just healthcare. The same way that banking-as-a-service (BaaS) and software-as-a-service (SaaS) have reshaped their respective industries, a new “as a service” model is emerging in the travel and tourism sector built on the back of AI capabilities.