ready or not, enterprises are betting on ai

AI Enterprise Deals and the Reality of Implementation

Recent weeks have witnessed a surge in agreements between artificial intelligence firms and large enterprises. Zendesk has introduced novel AI agents designed to resolve a substantial 80% of customer service inquiries. Simultaneously, Anthropic and IBM formalized a strategic alliance, and Deloitte announced its own partnership with Anthropic. Google also unveiled a new platform focused on AI solutions for businesses.

Challenges in AI Adoption

Despite these advancements, the integration of AI into established organizations isn't anticipated to be without obstacles. The timing of Deloitte’s announcement was notably coincidental, occurring on the same day Australia’s Department of Employment and Workplace Relations mandated a refund from the professional services firm. This was due to inaccuracies and apparent AI-generated fabrications within a report delivered to the department.

The latest episode of the Equity podcast featured Kirsten Korosec, Sean O’Kane, and a discussion of these recent AI developments. This was contrasted with prior coverage of the Sora application. While consumer-facing social networking applications powered by AI may eventually yield revenue, enterprise solutions offer a more immediate route to substantial financial gains.

Below is an edited excerpt of the conversation for enhanced clarity and conciseness.

Podcast Insights: Enterprise vs. Consumer AI

Anthony: I believe this situation underscores our previous conversation regarding generative AI-driven social networks. We considered these platforms as a potential revenue stream for AI companies, a prospect I still maintain. However, realizing this potential will require considerable time.

Enterprise solutions, though perhaps less glamorous than consumer applications, represent a more direct path to significant income. Sora may represent OpenAI’s monetization strategy in five years, but current revenue will be generated through enterprise contracts.

The Deloitte situation is particularly noteworthy. It’s a recurring reminder that these models aren’t consistently prepared for widespread deployment. However, it’s encouraging to see the Australian government challenge the report’s validity and demand accountability.

The use of AI in report creation isn’t inherently problematic, though a strong argument could be made against it. However, responsible oversight of the outputs is crucial. Information must be verified for accuracy, and simply submitting model-generated content for billable hours is unacceptable and warrants reprimand.

The Future of AI in Customer Service

Kirsten: Indeed. This week, Zendesk also announced new tools intended to automate a large portion of customer service interactions, potentially minimizing the need for human agents. Sean, are you observing similar automation trends in other sectors, such as the automotive industry and its service departments?

Sean: Yes, I’ve covered this topic extensively. Numerous startups are developing comprehensive customer service suites, including voice agents and large language models for handling email and text communications from dealerships and service centers. This is a promising development, as the primary issue isn’t a lack of available personnel, but rather the difficulty in reaching a representative or being transferred repeatedly.

Specifically within the service sector, customers often experience being redirected multiple times. Staff are frequently occupied. If AI can accurately capture inquiries and provide timely responses, it could significantly improve the customer experience. The key question is the extent to which businesses will embrace and maintain these technologies.

Many technologies, such as web forms, have been implemented by dealerships only to be neglected and rendered ineffective. Businesses often prefer direct phone calls. However, there is reason to believe that AI-powered solutions could become the initial point of contact for customers, and we are poised to discover whether this will become a reality.

Podcast Information

Equity is TechCrunch’s primary podcast, produced by Theresa Loconsolo, and is released every Wednesday and Friday.

Listen and subscribe on Apple Podcasts, Overcast, Spotify, and other podcast platforms. Follow Equity on X and Threads, at @EquityPod.