Generative AI (GenAI) has been at the center of conversations in the tech and business worlds for over a year, ever since the release of ChatGPT and the ensuing avalanche of investment, innovation, and product releases.
GenAI is fundamentally changing how we interact with data. It’s disrupting business functions and industries across the board. More and more technology platforms are embedding GenAI functionality. Like email and the Internet, it is becoming impossible to conduct business without an understanding of GenAI and the myriad products spawned from this technology.
As part of the 2024 Acceleration Economy Products to Watch series, I’m sharing the five GenAI products I’ll be watching most closely. Note: I’m not including OpenAI’s suite of tools because, in all likelihood, you probably already know them.
IBM watsonx & Partner Plus Program
Not too long ago, IBM released its new AI and data platform for the enterprise called watsonx. The platform includes three distinct components that collectively address enterprise requirements:
- watsonx.ai studio helps users build and deploy foundational GenAI and general machine learning (ML) models
- watsonx.data is a data store designed for AI built on an open lakehouse architecture
- watsonx.governance is a toolkit that helps customers build AI workflows responsibly and transparently
The watsonx suite is industry agnostic. Its earliest users include NASA and Wimbledon. IBM partnered with Hugging Face, a close runner-up for this list, to allow watsonx customers to build solutions with models hosted on Hugging Face.
To help you bring watsonx-powered solutions and automations to market, IBM also revamped its partner program, called Partner Plus. This is something we’ve unpacked in an interview with IBM and in the AI Ecosystem course. IBM’s Partner Plus program focuses on competitive incentives, insider access, education, and enhanced support and benefits for partners. The program also comes with access to IBM’s vast library of educational courses to upskill your workforce, which are the same materials provided to IBM employees.
Writesonic is a GenAI platform with several different products including Chatsonic, which helps produce written content needed for businesses; Botsonic, which provides plug-and-play AI chatbots; and multimedia products including Photosonic and Audiosonic. Unlike many other platforms, they support customers ranging from individuals to enterprises, helping customers to start with small implementations and scale.
Acceleration Economy chose Botsonic to help develop AnalystGPT, a private GPT instance that sources over 10,000 first-party reports, articles, and videos to answer AI, Cloud, Cybersecurity and Data questions.
“It has been a pleasure to co-create AnalystGPT with the team at Writesonic,” said John Siefert CEO of Acceleration Economy, “they clearly understood our mission to redefine the legacy tech analyst business by using GenAI to shorten the time to value for business technology decision makers trying to find trusted answers to big questions.”
“With AnalystGPT we have been able to move our subscribers from searching and clicking through our content, to asking questions and simply getting answers based on the analysis from our CXO practitioner analysts,” Siefert added. “Botsonic and the CIO at Acceleration Economy turned this vision into a reality through their co-creation of this GenAI solution.”
Implementation has gone smoothly, according to Acceleration Economy CIO Jose Copado, who pointed out, “Their onboarding has been awesome, their development team has been really open. They provide all the APIs. It’s pretty robust; you publish a piece of content, and it registers on their system, and you can see that they trained the model on it.” Bringing Botsonic onto the site itself was also easy. “The embed tools that they give you have been excellent, especially the ability to listen to the events of somebody clicking through the chatbot.”
Writesonic is a great platform for teams to rapidly test, implement, and scale a GenAI implementation. You can expect fair pricing, a friendly onboarding experience, and great customer service.
Last year, Databricks, the enterprise data management company, acquired a young company called MosaicML. With the integration of its product into the Databricks suite, I think MosaicML’s offering is one of the most competitive in the enterprise market.
MosaicML exists to help organizations train, build, and deploy ML models using their own data. Customers are offered two options: custom pre-training with their own data or fine-tuning of models that MosaicML pre-trained. MosaicML open-sourced these pre-trained models. The company understands enterprises’ unique privacy and customization requirements. Its training pipeline is straightforwardly priced on a compute basis, like any cloud provider. For deployment, it offers MosaicML-managed, on-premise, or cloud hosting.
As mentioned in our coverage of its acquisition, “MosaicML lets customers escape vendor lock-in, a common concern when working with proprietary foundational model providers like Meta, Google, OpenAI, and others. It also reduces the risk profile of working with a provider by giving the customer full control of the training process and by choosing not to store customer data…The flexibility of MosaicML’s platform also gives customers more control over their compliance and cost.”
And to add a cherry on top, Databricks’ acquisition of the company also brings the MosaicML training capabilities into Databricks data management suite. The result is an end-to-end solution for enterprises looking to build AI models.
Writer is an enterprise-focused GenAI platform. The key difference between Writer and other LLM providers like OpenAI is that it connects to actual business data in its so-called “Knowledge Graph.” This lets users generate collateral that’s on-brand, compliant, and contains fewer hallucinations. Customers can reference and tie responses to live data sources. It also offers self-hosting and has built robust guardrails into its product.
Writer trained its own models, which the company calls the Palmyra family. These are all open source, so the code and weightings can be inspected. Writer does not use customer data for retraining and all models are under 43 billion parameters, which means they’re powerful but not quite as clunky as GPT-4, minimizing cost and runtime for enterprise customers. The Writer user experience is clean and simple, letting anyone within your organization learn it quickly.
The company also confirmed that it used no copyrighted material or IP in its training data. That means if regulators crack down on copyright infringement in training data, then the company and its customers will remain safe.
Automation Anywhere Autopilot and Automation Copilot
Automation Anywhere is a leader in bringing intelligent automation to various functions within an organization. It’s unmatched in identifying opportunities for automation and putting them into place in your company, from HR to marketing to accounting and more. Its platform is underpinned by GenAI.
At its Imagine customer event last year, the company announced additional products that leverage GenAI: Autopilot and Automation Copilot for Automators. These tools allow for faster, streamlined automation development. For example, users can ask for a doc from Salesforce to be processed in ways A, B, and C, and the platform can automatically implement the sub-processes needed to make that happen. If code is needed, it will generate code.
Automation Anywhere also has unique dashboards for various teams in an organization so that handoffs are clean and cross-functional projects are manageable. Everything is backed up by real-world business data through direct integrations with enterprise applications including Salesforce, Microsoft D365, AWS, and more.
With Automation Anywhere, customers can quickly identify opportunities for automation and swiftly put them into place, without the hassle of juggling multiple teams, non-technical users, and scattered documentation.
We hope you enjoyed this list of our Top 5 GenAI products for business users. Without a doubt, the market is changing constantly, and this list will look very different at the end of the year. But until then, it’s worth exploring the above tools and evaluating whether or not your business can benefit from them.
More Analysis on Top Products for 2024: