Workbook 2

Applications of large language models

Explore the capabilities as well as potential and existing industrial applications of state-of-the-art large language models

2.1 How have large language models developed in recent years?

Over the past few years, the field of generative AI has made remarkable progress. In particular, these advances are reflected in both the significant increase in the size of Large Language Models (LLMs) and their increasing popularity (see figures below). These models have become increasingly versatile in understanding and generating different types of content, including text, images, video and source code. Recognising their significant potential, companies have begun to explore the use of LLMs in their products, services and processes across a wide range of industries, from e-commerce to healthcare, finance and more. In this way, large language models are changing how people think about AI and its ability to transform industries and society. 

Language model sizes in March 2023. Large language models have grown exponentially in size, driven by increases in computing power and the availability of high-quality training data. From LifeArchitect.ai.
The Hugging Face Hub, a popular open source platform for AI models, currently hosts over 60,000 models, 6,000 datasets and 6,000 demos. Notably, 2 of the top 3 most downloaded models are LLMs.

These recent developments lead to several important and impactful questions, such as: How can an organisation benefit from the use of large language models? What are some of the existing or potential use cases in products or internal processes where LLMs can have an impact? What types of results can be expected from such use cases? This chapter of the LLM Workbook aims to answer these questions by looking first at what large language models can do for industry, and then at existing or potential real-world applications of LLMs in specific industry sectors.

2.2 What can large language models do for industry?

Before exploring the use of large language models in real-world industrial applications, it’s important to have an understanding of several basic tasks that LLMs can perform in these use cases. These tasks, explained in turn below, are critical to the success of most industrial applications of LLMs. All the examples mentioned in this section, from a variety of start-ups and established companies, are publicly available.

  • Text Generation
    +
    –

    Text generation, also known as natural language generation, is the process of using input data, such as a sequence of words or phrases, to output text that is human-readable and coherent. It forms the basis of question-and-answer systems that can generate responses or answers to user input. Text generation has become one of the most prominent tasks in natural language processing, largely due to the popularity of ChatGPT, a large language model developed by OpenAI. Since its public release in November 2022, ChatGPT has demonstrated remarkable abilities to generate coherent and human-like text from a variety of inputs, attracting significant public and media attention. This has raised awareness of the potential of text generation in many industries, from manufacturing to software, where the generation of content such as articles, news, marketing slogans, social posts, minutes, customer feedback or product descriptions is an integral part of everyday work.

    With tools such as Copy.AI and Smart Copy, businesses can use the latest advances in natural language processing to automate the creation of virtually any type of content, including digital advertising copy, social media content, website copy, e-commerce copy, blog content and sales copy.

  • Image Generation
    +
    –

    Foundation models are now also capable of generating artificial images from pre-selected examples of images or text descriptions. Known as image synthesis, this has become an important application of generative AI models. In particular, the latest transformer-based models show significant improvements in generating photorealistic images with a wide range of content, style and variation based on text-based prompts.

    Examples of such models include DALL·E 2 from OpenAI and Stable Diffusion from stability.ai. DALL·E 2 is a 12-billion parameter version of GPT-3 that is trained to generate images from text descriptions, using a dataset of text-image pairs. Microsoft is integrating DALL·E 2 into its range of cloud AI services and software products, including the Microsoft Teams platform. Stable Diffusion, which was primarily developed in Germany, is a text-to-image diffusion model that has similar capabilities to DALL·E 2.

    Luminous, a family of large language models developed by Aleph Alpha, takes a different approach: Instead of generating images from text, its basic and advanced versions can read, explain and generate text based on existing images. This includes contextual Q&A, image captioning, and image analysis and summarisation. These features can be tested in the Aleph Alpha Playground.

    Significant advances in LLMs for image generation have led to a shift in the way AI-based content generation is viewed. Beyond pure text generation, various AI models are now focusing on multimodal generation of video, audio and even music.

  • Code Generation
    +
    –

    Code generation is the process of translating natural language into code or translating code between different programming languages. As large language models continue to evolve, code generation has become popular in the developer community. This technology allows developers to write code with minimal human input, making it a valuable tool for building applications more efficiently.

    By automatically generating code from natural language or other input, code generation technology can be used to enhance low-code/no-code apps, which have become increasingly popular in recent years. These tools allow developers to build applications without writing code by hand. Instead, they can use drag-and-drop interfaces to create workflows, forms and other features.

    While it is still unclear whether AI will replace the skills of software developers or require the development of new skills, natural language code generation shows promise as a tool that can help developers work faster and more efficiently. For example, CodeX by OpenAI can generate code from natural language input, making it easier for those who aren’t programming language experts to write software. Another example is GitHub Copilot, an AI assistant that helps developers by using machine learning algorithms to suggest snippets of code.

  • Machine Translation
    +
    –

    Machine translation has a long history, starting with rule-based solutions developed by IBM in 1954. Today, neural machine translation (NMT) is the most common approach, using deep learning to outperform statistical machine translation (SMT). German machine translation specialist DeepL has become one of the most popular machine translation services globally, with over a billion users. The popularity of machine translation continues to grow due to the ease with which it can be integrated into a wide range of end-user products,

    Integrated into NMT solutions, (multilingual) LLMs improve the quality of translation including in so-called low-resource languages, i.e. languages with only a small amount of digitally available corpora data (translation pairs) for training NMT models.

    Technology companies are investing heavily in research into different large language model architectures and methods optimised for machine translation. Amazon, for example, released the Alexa Teacher Model, which outperforms previous machine translation approaches and improves key services such as AWS Translate and Alexa. Alongside commercial options, open source projects like BLOOM produce coherent translations in 46 languages and 13 programming languages.

     

  • Text Summarisation
    +
    –

    Text summarisation, the process of reducing large amounts of text while retaining the essence of the original, is widely used in applications that summarise news articles or meeting notes, produce reports and briefings, and even generate mind maps from text. As the amount of information available increases exponentially in the digital age, there is a growing need to quickly sift through large amounts of text to identify key points. Large language models have emerged as the most effective choice for text summarisation tasks due to their superior language understanding capabilities for long and complex texts.

    The aforementioned Luminous family of LLMs from Aleph Alpha, available for testing on the Aleph Alpha Playground, is trained in the five most commonly spoken European languages (English, German, French, Italian and Spanish) and can summarise texts and documents in these languages.

  • Speech Recognition
    +
    –

    Speech recognition, also known as automatic speech recognition or speech-to-text, is the process of converting spoken language into text. This technology enables hands-free interaction with popular virtual assistants such as Siri, Alexa and Google Assistant, and is now ubiquitous in smartphones, smart speakers, cars and even robots. But speech recognition isn’t just limited to virtual assistants – it’s also widely used in call centres and dictation-related applications, such as supporting doctors in the healthcare sector or in manufacturing.

    Although speech recognition technology has made great strides in recent years, it has yet to achieve human-level performance. OpenAI has recently released a new automatic speech recognition (ASR) model, Whisper, that has the potential to change this. Whisper, which can be tried out on the Whisper Playground, demonstrates near-human accuracy and robustness, outperforming previous supervised and unsupervised models.

2.3 How are large language models used in industry?

Deploying large language models in practical, real-world industrial contexts can be challenging. Each industry has its own unique requirements, processes, vocabularies and jargon that need to be taken into account. It can also be challenging to find the right applications for large language models and to ensure that they perform well and produce robust results in these specific contexts. In this section, we look at real-world examples where large language models have already been successfully implemented with significant impact, as well as potential use cases where large language models could be useful. We cover a range of industries including healthcare, finance, manufacturing and more. Click on the interactive landscape below to navigate through the different industries and find examples relevant to your interests.

P aLM L uminous T5 W hisper M eg a t r on- T u r ing N L G S table Diffusion I magen video BE R T GP T -3 D ALL.E 2 C odeX Bloom T e x t G ene ra tion I mage G ene r a tion C ode G ene r a tion  T e x t summa r iz a tion Speech  R e c o g nition M achine t r ansl a tion T e x t I mage C ode A udio V ideo
Im Popup öffnen
P aLM L uminous T5 W hisper M eg a t r on- T u r ing N L G S table Diffusion I magen video BE R T GP T -3 D ALL.E 2 C odeX Bloom T e x t G ene ra tion I mage G ene r a tion C ode G ene r a tion  T e x t summa r iz a tion Speech  R e c o g nition M achine t r ansl a tion T e x t I mage C ode A udio V ideo

  • CUSTOMER GOODS & SERVICES

    Large language models have the potential to help businesses improve their products and services by better understanding their customers. Today’s customers expect smarter interactions. Providing more personalised, efficient and effective experiences can lead to increased customer loyalty and business growth.

    With LLMs such as ChatGPT, companies can more efficiently and automatically generate detailed product descriptions, offer customers intelligent image and voice search capabilities that go beyond simple keywords, and provide virtual assistants that deliver accurate and helpful answers. By using LLMs to analyse customer feedback, companies can identify common problems or complaints, tailor their offerings, differentiate themselves from competitors and protect their brand reputation.

    EleutherAI‘s GPT-J 6B engine, available on Amazon SageMaker, is an example of AI text generation. This technology provides businesses with valuable tools to improve the customer experience and their overall operations by speeding up and simplifying the implementation of customer service.

  • AUTOMOTIVE

    In the automotive industry, companies such as BMW and Tesla have demonstrated how the application of large language models can potentially revolutionise the industry, in particular by improving efficiency, customer satisfaction, innovation and growth. For example, LLMs can be used to optimise supply chains by analysing data from suppliers, manufacturers and distributors, identifying potential inefficiencies and making recommendations on how to streamline the supply chain and increase its efficiency. LLMs are also being used in voice-based in-car assistants, similar to popular personal assistants such as Siri or Alexa. However, greater accuracy and efficiency is required to meet stringent industry safety and regulatory requirements.

    It’s worth noting that compared to Chinese, American and Korean OEMs, most German carmakers tend to favour out-of-the-box options when choosing LLM solutions. BMW, for example, is using Amazon Alexa technology for its in-car voice assistant, while Nio, a well-known Chinese electric car manufacturer, deploys its own NOMI assistant.

  • FINANCIAL SERVICES

    Large language models promise to transform slow and complex financial reporting into real-time qualitative insights, acting as a kind of virtual CFO, always on hand to provide insightful financial guidance. By combining a deep understanding of global financial data with company-specific data, LLMs such as GPT-3 can, for example, show the cash flow impact of launching a new product, forecast future costs, determine the profitability of an acquisition and anticipate future expenses.

    AI in financial services is also set to have a significant impact on internal control, audit and regulatory compliance. It can identify anomalies such as discrepancies in invoices or discounts that don’t match contract terms, and alert management to suspected fraud and money laundering.

    The Tabulate application, for example, uses GPT-3 to provide a range of accounting and financial back-office services such as accounts payable, payroll, analytics, bookkeeping and tax preparation. With the help of LLMs, businesses are thus able to enhance their offering and provide high quality financial services to their customers.

  • SOFTWARE

    Code generation, which has a wide range of applications in IT processes or applications in any business, can be particularly important in software companies and has been shown to impact and drive business opportunities. For example, startup Debuild has developed an LLM-based application that generates SQL code to help developers build website applications and visual interfaces in a way that rivals human developers. At the re:invent 2022 conference, Amazon unveiled CodeWhisperer, which provides code recommendations based on developers’ natural language comments and code, aiming to increase developer productivity. Similar to GitHub Copilot, CodeWhisperer uses the context of comments and code to provide instant suggestions for individual lines of code as well as entire functions.

  • LAW

    Large language models deployed in the legal industry are already changing the way lawyers work, primarily by assisting rather than replacing them. LLMs are already being used to review contracts, find relevant documents during discovery and conduct legal research. More recently, they have been applied to new areas such as drafting contracts, predicting legal outcomes and even recommending bail or sentencing.

    In the US, law firms are already using LLM-based services from companies such as DISCO for discovery or Quick Check from Westlaw Edge to analyse draft arguments, aid legal research and improve the accuracy and efficiency of legal work.

  • MEDIA & COMMUNICATION

    As one of the first industries to adopt large language models, LLMs used in media and communications have already had a significant impact on the generation of news and journalistic content, for example, generating realistic and natural language explanations of current events in news organisations, or improving the quality and quantity of content in social media posts, blogs, vlogs or podcasts.

    The document editing tool HyperWrite, among many others in this area, allows users to create content in minutes and convert text into different styles and formats (e.g. short or long, informal or formal). The GPT-3-powered Grok tool can analyse and summarise discussions and threads in Slack, while Microsoft Teams Premium integrates ChatGPT to generate notes and tasks from meetings.

  • HEALTHCARE

    Capable of providing personalised health advice and easy-to-understand information for patients, large language models can be a useful tool for healthcare professionals. For example, PLANET AI‘s Doctor AI helps doctors make recommendations for standard tasks such as analysing medical images and processing medical documents such as lab data, clinical notes and other types of medical records. In China, AI doctor bots are treating patients in partnership with local governments and hospitals, and one-minute clinics have been set up to address shortages in rural areas, and local healthcare providers have been trained to use them.

    ai text gen

    In 2020, Chinese medical AI startups received $1.4 billion in funding, compared to $2.4 billion for their US counterparts, and the two countries accounted for 90% of global investment in medical AI startups. However, patient trust in “AI doctors” remains a concern.

  • ENTERTAINMENT & GAMING

    While the application of large language models to entertainment is still in its infancy, it clearly has the potential to transform how entertainment is experienced and enjoyed. The generation of natural language descriptions of various forms of entertainment, including movies, TV shows and games, can help media companies create accessible and engaging summaries of their content at a higher rate, thereby attracting more viewers and customers. In the games industry, using LLMs to generate storylines, images and even videos can help game developers create more immersive and engaging experiences. This can make games more exciting and appealing to a wider audience.

  • EDUCATION

    Large language models have the potential to revolutionise education and improve student learning outcomes, particularly by automating repetitive tasks such as marking student essays, providing personalised support such as feedback on student writing, and analysing data, freeing up teacher time and making learning more efficient and equitable. Virtual assistants based on LLMs can be developed to answer student questions. This provides real-time support and a more personalised learning experience.

    Squirrel AI, a Chinese online education technology provider with a focus on intelligent adaptive education, established a joint research center with Carnegie Mellon University in 2022 to explore personalised learning at scale and commercialise it internationally. With millions of students using “AI teachers”, China has demonstrated the potential of AI in education. In European countries such as Germany, where the COVID-19 pandemic has spurred the digitisation of education, AI could help overcome the shortage of teaching resources and further enhance the learning experience.

  • MANUFACTURING

    Manufacturing is a complex industry, with operations ranging from supply chain management to production, sales and after-sales support, made increasingly challenging by globalisation. LLM-based techniques have been used in manufacturing to automatically detect defects on the production line using computer vision, or to develop self-driving vehicles using image recognition and machine learning algorithms in the control system.

    While there are fewer mature use cases for LLMs in manufacturing than for image analysis, LLMs have great potential to improve efficiency and assist in prediction and maintenance. Proof-of-concepts (PoCs) have been tested in areas such as quality control, where LLMs predict defects based on production and planning data using text, images and even audio, or in intelligent maintenance, where smart assistants can improve efficiency and reduce costs.

  • MARKETING

    The use of large language models has had a significant impact on digital marketing. Among the many companies deploying GPT-3-based solutions in particular, MessageBird, which provides a unified platform for communicating across channels such as WhatsApp and Instagram, is integrating a GPT-3-based tool to streamline video meetings and authentication for marketing and sales for improved customer experience. Algolia, an enterprise search and discovery platform, leverages the power of LLMs to provide customers with easy access to product catalogues. Writesonic, a popular GPT-3-powered search engine optimisation (SEO) application, helps increase website traffic and visibility by generating SEO-optimised content for websites, including articles, emails, blogs and landing pages.

2.4 What’s next?

This chapter has explored the potential of large language models across a range of industries, including their capabilities and various real-world applications. We looked at the benefits that businesses can gain from using LLMs, as well as current and possible future LLM use cases in both products and internal processes. Large language models are set to become a core technology in human-machine or human-system interaction, with the potential to revolutionise a wide range of industries. However, their industrialisation is still a challenging task and much remains to be explored to extract the maximum value from this technology.

The next and final chapter of this LLM Workbook takes a closer look at potential risks and challenges associated with the current rapid proliferation of LLMs.