Unlocking Generative AI for the Enterprise

We are pleased to bring you a multi-part series where we will explore the impact of generative AI on various functions within the enterprise, such as marketing, engineering, customer support, and product. Our aim is to provide readers with an understanding of how generative AI is reshaping operations across businesses as well as actionable steps for embracing the technology's potential in their own organizations.

Artificial intelligence (AI) has undergone a remarkable transformation in recent years and is now evolving at a rapid pace. Not long ago, AI was primarily associated with automation technologies, including rule-based systems, expert systems, and machine learning algorithms. These earlier AI systems showed aptitude in analyzing and interpreting existing data for specific tasks, such as language translation, image and speech recognition, fraud detection, and predictive maintenance. Today, we stand at the cusp of a new era in AI - the era of generative AI. This paradigm shift promises to revolutionize the enterprise landscape by enabling AI systems to create novel content spanning images, videos, and text, with significant implications for innovation and human productivity.

Generative AI has taken center stage since last year, following the public launch of machine learning models such as ChatGPT, Stable Diffusion, and DALL-E 2. The adoption of these applications has skyrocketed – ChatGPT, for instance, amassed 100 million active users within just two months of its November 2022 debut, setting a new record as the fastest-growing consumer app in history. This remarkable success has inspired a burgeoning wave of startups eager to harness the potential of these advanced models and tailor their applications for specific industries, such as healthcare, media, consumer brands, fintech, and enterprise software. According to? Pitchbook, investment in generative AI companies has already surpassed $3B in 2023, signaling widespread developer interest coupled with real-world (and relevant) use cases for the technology.

While the application layer of generative AI is still nascent, the underlying technology has been evolving for years. The current landscape of generative AI has been shaped by a series of cumulative advancements in models, data, and compute power.

Let’s delve deeper into some of these pivotal moments to better understand their impact and significance:

1. Generative AI gained prominence in 2014 with the debut of generative adversarial networks (GANs), a deep learning framework featuring two rival neural networks that generate realistic outputs like images.
2. The 2017 debut of the transformer model revolutionized natural language processing (NLP) tasks by learning context from entire input sequences, not just sequential data. These models reduced reliance on large labeled datasets and opened up abundant online data, given their self-supervised learning ability.
3. In 2021, OpenAI introduced Contrastive Language-Image Pre-training (CLIP), a technique for text-to-image generation. When combined with diffusion, a deep learning technique for generating images from noise, CLIP classifies data based on the likelihood of matching a text prompt. This technology underpins OpenAI's DALLE-2 and open-source alternatives like Stable Diffusion, enabling them to generate high-resolution, detailed images.
4. As generative models advanced, so did computational infrastructure and platforms. Various platforms now simplify running models at scale. Product teams can use open-source models, like BLOOM and Stable Diffusion, or access solutions like OpenAI API and HuggingFace Inference Endpoints to bypass in-house ML expertise.

Generative AI has immense potential for enterprises, particularly when it comes to enhancing efficiency and unlocking cost savings at scale. Brett Wilson, co-founder and General Partner of Swift Ventures asserts that "today's generative AI applications have just scratched the surface…organizations will increasingly use a combination of models, including their own to leverage their now private and real-time data sets.” Goldman Sachs Research predicts that generative AI could contribute to a 7% increase in global GDP and boost productivity growth by 1.5 percentage points over a decade. OpenAI's research further suggests that generative AI can significantly impact the workforce by automating labor-intensive tasks, leading to increased productivity and global economic growth.

While predictions like those above regarding generative AI’s potential are increasingly common, it is important to weigh potential risks and uncertainties that may impact the opportunity for startups and adoption:

• Centralization of AI models: The largest tech companies could exert significant influence on the generative AI landscape, as their extensive resources, data access, and talent position them to develop, train, and refine AI models more effectively than smaller players. This dynamic could limit emerging companies to the application layer, where they would build solutions atop AI models developed by these larger tech companies.
• Adoption timeline: Concerns persist about the timeline for broader generative AI adoption. Despite rapid technological advancements, integrating AI into operations may take longer than expected for enterprises and startups.
• Data security and privacy:? As businesses increasingly rely on vast amounts of data to train AI models, risks of misuse or unauthorized access to sensitive information rise. Robust security protocols and data governance frameworks are essential to mitigate these risks and ensure the responsible use of generative AI technologies.
• Reliability: Generative AI models may produce unexpected or inaccurate outputs, particularly in novel situations or with unfamiliar data. Businesses should carefully evaluate the technology's applicability and limitations within their specific use cases.

As generative AI adoption continues amongst employees despite these risks, it is vital for executives to stay informed of emerging trends and use cases to guide their organizations toward success. As evidenced by the following quotes, CEOs already recognize the transformative potential of generative AI for business.

As we began forming our thesis on Generative AI, we conducted extensive research and spoke with a range of experts, including senior executives, startup founders, and other fund managers. Through our efforts, we gained valuable insights into how large enterprises are approaching generative AI, the pain points they are addressing, and the competitive landscape of startups in the space.

Throughout this series, we will showcase real-world examples of organizations embracing generative AI, highlight startups that are leading the way, and discuss potential risks and challenges that enterprises may face during the adoption process. So, stay tuned as we embark on this exciting journey into the world of generative AI and its powerful implications for the enterprise.

Sources:

Hu, Krystal. “ChatGPT Sets Record for Fastest-Growing User Base - Analyst Note.” Reuters, February 2, 2023. https://www.reuters.com/technology/chatgpt-sets-record-fastest-growing-user-base-analyst-note-2023-02-01/.

?“PitchBook,” n.d. https://my.pitchbook.com/search -results / s275397515 / deal_chart.

Goldman Sachs. “Generative AI Could Raise Global GDP by 7%,” n.d. https://www.goldmansachs.com/insights/pages/generative-ai-could-raise-global-gdp-by-7-percent.html.

“GPTs Are GPTs: An Early Look at the Labor Market Impact Potential of Large Language Models,” n.d. https://openai.com/research/gpts-are-gpts.