In this article:
- What is generative AI?
- 3 key players in LLM-based generative AI
- How organizations and employees can leverage the power of generative AI
- Addressing employee fears around AI
- Ethical considerations of implementing AI systems
- Navigating generative AI as an adult student at Penn
- Ready to pursue an Ivy League education?
The widespread availability of generative AI applications has the potential to dramatically alter the way we live and work. According to research from The Future of Jobs Report 2023 by the World Economic Forum, more than 75 percent of companies plan on adopting big data, cloud computing, and artificial intelligence (AI) in the next five years.
This comprehensive guide provides an overview of generative AI, explains how it can drive operational efficiency when implemented judicially, and addresses employee concerns and ethical considerations regarding its widespread adoption.
What is generative AI?
Generative AI is artificial intelligence that uses algorithms to create new, realistic content such as images, text, audio, code, video, and simulations in response to natural language requests or other prompts. Generative AI models are “trained” on large, complex sets of existing data. They can use this knowledge to produce outputs, including everything from articles, marketing copy, and film scripts to architectural renderings, automotive designs, and weather forecasts.
According to data from McKinsey’s The State of AI in Early 2024, the adoption of AI technology has more than doubled since 2017, and the top reason for doing so is to optimize service operations.
Three generative AI applications currently used across organizations include:
- Natural language text – Used for translating; writing code; grammatical correction and analysis; and business, academic, and creative writing
- Audio and speech – Implemented for dubbing, composing, and songwriting; speech and voice recognition; sound editing; and dictation and transcription
- Visual and imagery – Employed in 3D modeling, illustration, infographics, video, creative and product design, architectural rendering, and image editing
Summary of the science behind generative AI
The science behind how generative AI works is complex. This technology is a form of machine learning that involves training computer software to make predictions based on data.
Massive amounts of existing content are inputted into AI models (or algorithms) to teach them how to generate new content or other outputs. These models learn how to identify patterns and relationships grounded on a probability distribution and then develop similar patterns when provided with new requests.
Deep learning is a machine learning subgroup inspired by the human brain, wherein generative AI uses a neural network to manage more complex patterns than those traditionally tackled through machine learning.
Generative AI tools use diverse models to train the artificial intelligence and deliver outputs, including transformers, variational autoencoders (VAEs), and generative adversarial networks (GANs).
What is an LLM?
A large language model (LLM) is an AI model that allows chatbots, such as ChatGPT, to respond to text-based prompts or inputs. Unlike other forms of generative AI that may produce audio, images, or videos, LLMs are specifically designed for tasks related to natural language generation and comprehension.
Training data from existing source materials allows an LLM to identify relationships between words and phrases, understand rules of grammar and semantics, and generate coherent and contextually appropriate responses to different requests.
3 key players in LLM-based generative AI
A variety of LLM-based generative AI tools are available to help save time and increase operational and individual efficacy in the workplace. Below are three popular options.
1. OpenAI’s ChatGPT
ChatGPT is a free, AI-powered chatbot that allows users to ask questions, engage in conversations, challenge inaccurate premises, and prompt it to compose different forms of text, such as articles, poems, or summaries of information. OpenAI also offers a subscription plan for ChatGPT Plus with faster response times and priority access to new features and upgrades.
2. Open AI’s DALL-E
DALL-E is a text-to-image generative AI interface trained to create new, realistic images and art from natural language prompts. DALL-E can combine different styles, attributes, and concepts in its images. OpenAI released a second iteration of the interface in 2022 to offer more realistic and accurate image generation at a much greater resolution.
3. Google’s Gemini
Gemini is a text-to-text generative AI tool designed to help users experiment with creative ideas and explain concepts. It can create text, translate different languages, and produce content based on prompts. As is the case with other generative AI, Bard has the potential to provide inaccurate information, so Google recommends double-checking its responses.
How organizations and employees can leverage the power of generative AI
When implemented with transparency and communication in mind, generative AI tools can bolster task efficiency, simplify content creation, develop new products, and more across industries.
Generative AI’s impact on automation and operational efficiency
One of the key benefits of generative AI is its ability to automate repetitive tasks, which can streamline workflows, reduce business costs, and enhance productivity. In addition, data-driven insights from AI can help improve decision-making, pinpoint bottlenecks, identify revenue opportunities, and provide a competitive advantage.
According to a recent report by McKinsey Digital, generative AI can provide value for organizations by producing personalized content for sales and marketing, improving customer service experiences, facilitating faster software programming and coding, and reducing research and development (R&D) time.
Automating certain individual responsibilities can also free up time for employees to tackle more strategic, high-level tasks. This has the potential to boost overall labor productivity significantly, provided that organizations across industries invest in skills training and support as employees take on different roles and responsibilities.
Additional key findings from McKinsey Digital’s research include the following:
- Generative AI’s impact on productivity could add anywhere from $2.6 to $4.4 trillion annually to the global economy
- Approximately 75 percent of that potential value falls into four areas: marketing and sales, customer operations, software engineering, and R&D
- Banking, life sciences, and high tech could see the largest impact as a percentage of revenues from generative AI
- Current generative AI technologies could automate activities that take up 60 to 70 percent of employees’ time today
- Given the potential for technical automation, 50 percent of today’s work tasks could be automated between 2030 and 2060
How employees can use AI to their benefit
When employees embrace generative AI tools like ChatGPT to generate content, simplify tasks, and improve efficiency, it can benefit individuals and organizations alike. Examples of practical ways employees can use AI to streamline daily tasks include:
- Writing drafts for articles, emails, or marketing materials
- Generating outlines for business proposals, reports, or plans
- Developing summaries of business intelligence or research
- Professionalizing the tone of business communications
- Creating photorealistic art to accompany presentations
- Using chatbots to manage customer service and technical or general requests
Industry-specific generative AI applications to improve performance also abound:
- Doctors and practitioners can use AI to analyze medical images and aid in diagnoses
- Teachers can use chatbot tutoring and generative AI tools to help create lesson plans
- Journalists and creatives can leverage AI to summarize, generate, and edit content, images, audio, and video
- Scientists can implement technologies to simulate natural disasters and model climate scenarios
- Lawyers and paralegals can use AI to suggest legal arguments, analyze evidence, and design and summarize contracts
- Vehicle manufacturers can run models for automotive simulations or test new product developments
Addressing employee fears around AI
Although AI offers many benefits, there is still fear that AI could replace human workers in the near future. That’s why organizations should communicate to employees that their unique skills and capabilities are essential to successfully integrating generative AI applications into the workforce.
Job security
Job security is a genuine concern for people. According to predictions by Gartner, by 2025, more than 30 percent of drug discovery will be via generative AI; 30 percent of external marketing messages from large companies will be AI-generated; and generative AI in automotive, manufacturing, aerospace, and defense will be used to accelerate and optimize the design process.
Per Gartner insights, industries most likely to be affected by AI include architecture, automotive, aerospace, defense, electronics, manufacturing, media, design, marketing, corporate communications, medicine, pharmaceuticals, and software engineering.
However, the outlook for job growth due to technological advancement is quite promising. According to data from The Future of Jobs Report 2023:
- The effect of most technologies on jobs is predicted to be a net positive over the next five years—with climate change and environmental management technologies, big data analytics, encryption, and cybersecurity driving this growth
- Most of the fastest-growing roles are driven by digitalization, technology, and sustainability—including AI and machine learning specialists, sustainability specialists, business intelligence analysts, information security analysts, and renewable energy engineers
- The adoption of new and frontier technologies and increased digital access are predicted to propel job growth in more than 50 percent of companies surveyed, with 20 percent expecting job displacement
Balancing tech advancements and human roles
Artificial intelligence and human intelligence offer distinct and complementary strengths. The former can automatically perform repetitive tasks, generate content, and analyze massive amounts of data, while the latter can infuse empathy, creativity, communication, and critical thinking skills.
According to the World Economic Forum’s report, companies surveyed view analytical and creative thinking as the most crucial skills for employees in 2023. Resilience, flexibility, agility, motivation, self-awareness, curiosity, and lifelong learning are valued as well. Employers predict that skill disruption will affect 44 percent of employees in the next five years.
Workers who want to remain competitive and spur career growth may benefit from educating themselves on the latest AI technologies while prioritizing upskilling in the areas above. Organizations implementing AI systems should provide professional development and training programs that allow employees to enhance their skills in these areas.
Although not all of the following examples employ generative AI or large language models specifically, these companies are enjoying promising human-AI collaborations:
- HubSpot implements AI tools to analyze customer data and help businesses target and customize their marketing and advertising campaigns
- John Deere uses AI and machine learning to analyze soil samples and help farmers make data-driven crop decisions, such as where and when to plant
- Airbus created AI to predict when parts may need to be serviced or replaced, which helps reduce maintenance costs and improve safety
- JPMorgan Chase developed COIN, an AI system designed to help bankers process loan agreements more swiftly and precisely
- The University of California, San Francisco (UCSF) created AI technology to help radiologists analyze mammograms and identify potential areas of concern
Ethical considerations of implementing AI systems
Although generative AI offers many promising benefits to organizations, it brings real ethical concerns that are rapidly evolving with the technology.
Current generative AI tools can reproduce human language so coherently that it can be difficult to discern whether or not AI has produced the content or determine if it is false or misleading. AI can also be used for harm in the wrong hands. Because it can generate convincing voice recordings, photorealistic images, and video, artificial intelligence can enable complicated scams and deep fakes or stoke political division.
Additional oversight risks and recommendations for addressing them include:
- Accuracy – Assess all outputs for inaccurate or fabricated answers and the level of appropriateness before distributing content
- Transparency – Clearly communicate and label generative AI content for public consumption
- Bias – Put policies into place to analyze outputs and ensure that prejudicial views do not compromise them
- Cybersecurity/fraud – Prepare for AI-generated cyber and fraud attacks targeted toward employees, clients, or customers
- Intellectual property and copyright – Assume that all queries or data entered into generative AI tools will become public information. Remember that content created by ChatGPT and the like is trained on vast sets of publicly available data and not compliant with General Data Protection (GDPR) and other copyright laws.
- Sustainability – Filter and prioritize use cases for AI in your workplace to lessen power consumption, and utilize renewable energy to offset the significant amounts of energy required for generative AI
Navigating generative AI as an adult student at Penn
As a world-class research institution, the University of Pennsylvania embraces innovations in technology including artificial intelligence. To help the Penn community ethically and responsibly explore the educational applications of this rapidly evolving technology, Penn Information Systems and Computing released the following guidance for students:
- All use of AI should be in line with Penn’s Code of Student Conduct and the Code of Academic Integrity.
- Educators may have requirements and guidance for citing the use of generative AI output and for attributing AI created content to the specific AI tool and parameters used.
- Individual courses may have different or more narrow guidance on the use of AI that should be adhered to within the context of that course.
- In the absence of other guidance, treat the use of AI as you would treat assistance from another person. For example, this means if it is unacceptable to have another person substantially complete a task like writing an essay, it is also unacceptable to have AI to complete the task.
- Keep in mind that having access to data is not the same as having permission to scrape the data or use it to train an AI model.
In other words, Penn LPS Online students should refer to their course instructor and syllabus for individual course guidelines on whether and how generative AI may be used for coursework. When generative AI is permitted for a course, it must be cited as a source.
Ready to pursue an Ivy League education?
The Bachelor of Applied Arts and Sciences is a flexible and accessible liberal arts degree for working adults and nontraditional students. With a variety of online Ivy League courses and degree concentrations to choose from—including Data Analytics and Psychological Sciences and Leadership and Communication—you can customize your education according to your personal and professional goals.
Penn LPS Online certificates are short programs designed to enhance specific skill sets, such as digital strategies or data analytics. Whether your goal is pursuing additional learning, a job transition, or career advancement, these credentials are an ideal alternative to completing a bachelor’s degree.
If you haven’t already, apply to Penn LPS Online today. View our course guide to learn more about what’s available in any upcoming term.
