Revolutionizing Work Productivity with Generative AI — Part 2

Building enterprise knowledge bases and planning for a connected enterprise that leverages AGI

Connected Enterprise — AI + Automation

Building Enterprise Knowledge Base with Generative AI

The public domain Gen-AI use cases are limited to the period of the data as well as access to open datasets (as shown in part 1 of this series). To overcome this, enterprises can create their knowledge bases using customer tickets, surveys, and internal HR and company information by aggregating, cleaning, and preprocessing the data.

Then, use this knowledge base as input to train generative AI models. This will enable the AI models to learn from various internal datasets, thus overcoming the limitations of public domain Gen-AI use cases.

Please note that while handling sensitive data like customer information or HR data, ensure proper data security and privacy measures are in place, following your organization’s policies and relevant data protection regulations.

The use cases below are examples of responses generated and references on how the answers were derived. This will help build trust in such models since the responses from such assistants are direct answers, as opposed to links to other sources. Additionally, the references provided with the answers offer transparency into how the model arrived at the response, further increasing user confidence.

💡 Q & A from HR policy documents

Building an enterprise knowledge base using generative AI

Quickly and accurately generating responses is particularly useful in customer service and technical support, where quick and precise answers are often crucial to keeping customers satisfied.

Companies working on generative AI use cases can protect intellectual property (IP), privacy, GDPR compliance, and personally identifiable information (PII) data by implementing several strategies.

These strategies include the following:

1. Data anonymization: Anonymize data by removing personally identifiable information (PII) such as names, addresses, and social security numbers. This process can help ensure GDPR compliance and protect privacy.
2. Data encryption: Encrypt data both in transit and at rest. This provides an additional layer of security to protect sensitive information from unauthorized access.
3. Access control and authentication: Implement strict access control and authentication measures to ensure that only authorized personnel can access sensitive information. This includes using multi-factor authentication and role-based access control.
4. Data minimization: Collect and process only the minimum amount of data necessary for the specific purpose. This principle is a core requirement of GDPR.
5. Data retention and deletion policies: Establish clear data retention and deletion policies to ensure that personal data is not stored longer than necessary. Regularly review and update these policies to maintain compliance with GDPR and other privacy regulations.
6. Privacy by design: Integrate privacy considerations into designing and developing AI systems. This includes implementing data protection measures at every system lifecycle stage.
7. Regular audits and assessments: Conduct regular audits and assessments to identify and address potential vulnerabilities in AI systems. This may include privacy impact assessments (PIAs) and data protection impact assessments (DPIAs) as GDPR requires.
8. Employee training and awareness: Provide comprehensive training on data protection and privacy best practices. This helps ensure all team members know their responsibilities when handling sensitive information.
9. Third-party risk management: Assess and manage the risks associated with third-party vendors and partners, ensuring they adhere to data protection and privacy standards.
10. Legal agreements and IP protection: This area is a work in progress for derivative work as NDAs and general IP protection can not be easily enforced.

Future of AGI and Auto GPT for Enterprises

The future of generative AI in enterprises is exciting.

As companies continue incorporating AI and automation into their workflows, generative AI will play an increasingly important role in creating content and automating processes. Auto GPT, which stands for Automatic Generative Pre-training Transformer, is a technology that allows for even more advanced levels of workflow orchestration.

Imagine a future in which the prompts listed above can be chained together in a workflow, and with human input required only for validations and approvals, sophisticated tasks can be completed while recording a history of all the steps and the rationale behind the decision steps.

The example below is a concept that provides an end goal by defining a list of tasks and then triggering bots to complete these tasks before generating additional ones based on the finished items.

I am a new hire in my company. Use SAP Ariba to buy me a laptop, and SAP Concur to book my onboarding trip from Philadelphia to San Franciso. My laptop budget is $1800 and my trip budget is $2500. I would prefer a nonstop flight and Marriott as the preferred hotel to stay.

Building a connected enterprise using AI + Automation

Other compelling use cases include:

Headcount planning — I have a skill gap in machine learning experts with 5–10 years of experience and enterprise software knowledge. Use my budget of $2.5M for the year to recommend a hiring plan that combines at least 25% full-time employees and should be run 70% from a low-cost location.

Win/Loss — Analyze the customer support tickets for all accounts over $5M in annual revenue that is not renewing next year. Provide a detailed breakdown of customer interactions, including customer support activities, ongoing deal cycles, as well as any account or success manager turnovers.

It is expected that Auto GPT will revolutionize how content is produced and distributed in the coming years.

Conclusion

AI + Automation Timeline

In this two-part(part 1) series blog, we looked at how enterprises can:

1. Provide access to the right prompts for teams to help them be more productive and creative and increase the quality of their skills.

Building an intelligent connected enterprise using AI + Automation

2. Enrich the data set by providing access to internal knowledge bases that can help employees have faster access to information and equip them with real-time competitive intelligence.

3. Think about a combination of AI + automation use cases(AGI) that can focus on completing tasks at hand by combining multiple steps, systems, and workflows with minimal human intervention.

If you have any feedback or corrections to the blog or want to test drive the co-pilot app, please contact me.