Since last year's launch of ChatGPT, Artificial Intelligence (AI) has dominated news cycles in 2023. AI's reign as the leading technological innovation is expected to continue into 2024. The focus for enterprises will be on remaining vigilant and adopting a skeptical eye, separating fact from fiction.
2023 was a breakout year for the adoption of AI tools. It seems like every enterprise function is now actively experimenting with AI and its subsets such as Generative AI (GenAI), and integrating it into daily work processes. When looking at the promised benefits, who can blame them? Increased end-to-end efficiency, improved accuracy and decision-making, a reduction in costs, more empowerment for employees, a superior customer experience — the competitive advantages for businesses seem endless.
The stakes are high for leaders who want to incorporate AI into their businesses: three out of four executives understand they need to scale AI across the organization to stay competitive—and in business entirely.
Early AI adopters need to keep a critical eye on these AI-related developments that are happening at break-neck speed. That means asking tough questions about AI and the associated tools, solutions, or vendors you’re planning to work with.
* that record has since been dethroned by Threads
One year on from the launch of ChatGPT, as with any hype cycle, the initial wave of excitement has gradually subsided, giving way to a more critical evaluation of ChatGPT's merits and potential shortcomings. The big question is whether the hype surrounding ChatGPT and AI is indeed warranted.
Although AI is clearly in its infancy, several emerging factors provide compelling evidence to suggest a shift in AI's trajectory and its potential to stick around this time:
The confluence of these factors signals that AI is approaching a critical juncture, poised to transition from experimentation to widespread adoption. The future of AI seems bright, and we can expect to see it revolutionize many aspects of our lives.
Specifically for the order-to-cash cycle and accounts receivable, where the onus is on moving and settling funds faster than ever before, AI offers particularly sharpened opportunities. As Nick Izquierdo, Executive Vice President of Payments at Billtrust, recently pointed out in a Pymnts article, AI has the potential to really move the needle for so much of the industry in terms of the ability to better understand what’s going on within payments environments and the ability to be more proactive with buyers or suppliers.
The rapid advancement of AI has fueled a surge of excitement, but it's crucial to temper expectations to avoid disillusionment. Gartner's hype cycle aptly describes this pattern of anticipation followed by disappointment.
With the AI hype reaching new heights in 2023, businesses face the challenge of discerning whether an AI solution is poised for success or destined for disillusionment. A healthy dose of skepticism is essential to separate hype from reality and make informed decisions.
Adding to the challenge is the tendency of some companies to exaggerate AI's role in their products. While these products may still be valuable, they may not be truly powered by AI. Thorough research and detailed demos from potential vendors are crucial to cut through this hype.
Absent a strong counter-narrative to the dominant tech ideology around AI, we are likely to witness multiple AI-branded technology bubbles emerge in the coming years. Navigating this landscape requires a discerning eye and fact-based decision-making.
To avoid being fooled by AI hype, it's essential to approach AI with a discerning eye. Be wary of exaggerated claims and unrealistic expectations, and conduct thorough research. By understanding AI's capabilities and limitations, you can unlock its true potential while avoiding the pitfalls of overhype.
AI hype often manifests in several ways:
Maintaining a healthy dose of skepticism is key. Today’s tools can be used to improve efficiency, but they cannot replace human judgment. Many tasks, such as making financial decisions, still require human judgment, empathy, and creativity - all of which AI cannot replicate on its own. In short, AI should complement human capabilities, not replace them, leading to better decision-making and outcomes.
While many AR vendors are jumping on the AI bandwagon, it takes time - and a lot of thought - to develop a good and truly vetted AI solution and deliver tangible value. Data needs to be present in the right state, models must be established and trained. Anyone claiming to stand one up without clear diligence is not doing it right.
At Billtrust, we have an integrated approach and strategy to implementing AI in our products. We take a lot into consideration:
Compatibility
Compatibility is the foundation of a successful integration, and the AI in our solutions aligns seamlessly with your existing systems and processes. It’s not a separate product but always works on top of your existing AR data cloud.
Scalability
Solutions that target individual workstreams may struggle to accommodate growing transaction volumes or evolving business needs, limiting scalability and adaptability. Our SaaS offerings can grow with your business. The confidence-based machine learning (ML) in our Cash Application solution doesn't require programming to update rules; more volume doesn’t mean more work.
Ease of integration
Addressing your specific needs requires harvesting data from across your organization and potentially beyond. A major challenge lies in seamlessly integrating data from disparate systems, a common hurdle for IT departments. Our approach prioritizes minimizing disruptions to your workflow when connecting with your financial data repositories.
Enriching your proprietary data with external sources, including public data, AI tools, and third-party data providers, can significantly enhance the AI's ability to grasp your context, anticipate your requests, and draw from a broader pool for command execution. However, proper integration is crucial. The adage “garbage in, garbage out” still holds true when integration is not done correctly.
Support
Robust customer support and comprehensive training are non-negotiables at Billtrust. We gladly offer support to help your team navigate the AI landscape effectively.
Security
We take data security into account in everything we do. Our secure data lakehouse functions as a single source of truth, and fulfills a variety of functions. One of them is forming the basis for future advancements with generative AI. Part of that is making sure your proprietary financial data stays secure when it’s used in these LLMs. Read more about Billtrust’s data security measures.
ROI matters
We try, first and foremost, to comprehend your workflow and the challenges you’re encountering. Only then can enhancements be made, and you can achieve any meaningful returns on investment. The key also lies in consistently driving year-over-year improvements.
Unlike data science models making precise predictions, large-language models like ChatGPT provide persuasive responses, potentially inaccurate ones. For content ideation, this is acceptable, but for financial data, accuracy is paramount. To mitigate risks and make an interesting business use case, you’ll need to create a framework around these large-language models that can give you more deterministic and accurate answers by validating and verifying outputs.
Once a framework is in place, you can interact with your accounts receivable and other financial information in a natural and intuitive way. By posing questions in a conversational format - utilizing a ChatGPT-like interface - you can retrieve the data you require from your source systems.
The data you need varies depending on your role. Billtrust understands the importance of this. As a decision-maker, you want predictive data to inform future planning. If you’re in a more hands-on operational role you’d want to be presented with data that allows you to do your specialist job much faster and efficiently. Ideally, you have direct and independent access to get you the information and data you need to do your job, whether it is that of a CFO or a collector.
Billtrust’s comprehensive AI and analytics approach encompasses three key pillars:
Decision analytics: Transactional reporting and analytics modules to show basic KPIs and help you make business decisions.
Predictive analytics: Detailed reports and metrics that allow you to take proactive action on your data.
Generative AI: Prescriptive insights and cognitive workflows to provide benchmarks, answer questions, and provide recommendations.
Furthermore, we recognize that autonomous analytics are just one component of a comprehensive AI strategy. Our focus at Billtrust is to set the foundation for future innovation with scalable features that will revolutionize AR for years to come, while keeping data security and privacy top of mind by employing robust security protocols and adhering to strict data privacy regulations.
If you have questions or need more personalized guidance, please don't hesitate to reach out. Your success remains our top priority.