Table of Contents
1. Preface
2. Business Value
3. Smart Factory
4. MLaaS Market
5. Smart Workflow
6. Generative AI
7. Blueprint for AI Bill of Rights
8. Andrew Ng Predicts the Next 10 years in AI
9. Model Hubs
10. Conclusions
11. About Us
12. References
13. Contributors
Preface
As the use of machine learning (ML) and artificial intelligence (AI) in business becomes more prevalent, it is increasingly important for organizations to understand its potential value and the readiness of companies to adopt and implement it. In this report, we delve into the current state of ML in the business world, examining the trends, challenges, and best practices that are shaping its adoption and use.
We begin by exploring the various ways in which ML is being leveraged to drive value for businesses, including increased efficiency, improved decision-making, and the development of new products and services. We also take a closer look at the hot areas of ML and technology, such as generative AI and smart factories, and examine the current landscape of investments in this technology.
In addition to these broad trends, we also delve into the emergence of model hubs, which are tailored to specific industries and offer unique value for businesses operating within those sectors. Finally, we discuss the trends that we are seeing in the ML space based on our own experiences, offering insights and recommendations for organizations looking to capitalize on this rapidly evolving technology.
This report provides a comprehensive overview of the current state of ML in business and offers valuable insights and guidance for organizations looking to harness its potential.
Business Value
ML and AI have the potential to bring significant value to businesses by enabling them to automate tasks, make more informed decisions, and improve their products and services. However, in order for businesses to realize this value, they must be ready to adopt and integrate these technologies into their operations.
The readiness of companies to adopt ML/AI varies widely. Some companies are already well-established in their use of these technologies, while others are just starting to explore their potential. In general, companies that are most ready to adopt ML tend to have strong leadership and a culture that is open to innovation and change. They also tend to have the necessary resources, including skilled personnel and access to data, to support the development and deployment of these technologies.
In order for companies to realize the full value of ML, it is important that they take a strategic approach to their adoption and integration. This may involve investing in the necessary resources, building internal capabilities, and developing partnerships with external experts. By taking a proactive and well-planned approach, companies can position themselves to benefit from the transformative power of these technologies and stay competitive in an increasingly digital world.
It is difficult to provide specific statistics on the business value of ML and the readiness of companies to adopt these technologies, as these will vary widely depending on the specific industry, company, and use case.
A recent study done by MIT Sloan showed that there is a significant difference between ambition and actual implementation of AI in most companies. While a majority of executives (75%) think that AI will enable their firms to expand into new areas of business and a large portion (85%) believe it will give them a competitive edge, only a small fraction (5%) have extensively integrated AI into their products or processes. Even fewer companies (20%) have even partially adopted AI. Furthermore, only 39% of all companies have a strategy for implementing AI. Among larger companies (those with at least 100,000 employees), the adoption rate is slightly higher, but still only half have an AI strategy in place [1].
However, there is a growing body of research and data that suggests that AI and ML can bring significant value to businesses. The use of AI in the Banking, Financial Services, and Insurance (BFSI) sector is expected to see significant growth in the coming years. In 2022, the global market for AI in BFSI is estimated to be worth $3232.9 million, but it is projected to reach $15320 million by 2028, a compound annual growth rate (CAGR) of 29.6%. This growth is due in part to the impact of the COVID-19 pandemic and recent events over the last few years [2].
These statistics suggest that there is significant potential for AI and ML to bring value to businesses, and that many companies are already starting to adopt these technologies. However, it is important to note that the actual business value and readiness of any given company will depend on a variety of factors, including the specific use case, the quality of the data, and the availability of skilled personnel.
Smart Factory
Furthermore, the vast potential of AI and ML has been realized in the form of smart factories, which are the next step in advancing the manufacturing industry. What sets these smart factories apart from their predecessors is the fact that they can operate automatically and can adapt with the latest technology, allowing them to constantly improve all aspects of production.
Currently, smart factories market size are valued at USD 129.74 billion in 2022 and are projected to grow to USD 321.98 billion by 2032 (growing at a CAGR of 9.52%). The reason for their explosive projected growth is mostly attributed to the advancement of IoT (Internet of Things), cloud computing, and AI. Extensive utilization of manufacturing execution systems (MES) and sophisticated data models for process-specific operation [3]. On top of the plethora of technological advancements, sensors integrated into the system record vital data that will help businesses optimize and smoothen their day-to-day operation.
Businesses that are looking to optimize their production should adopt the smart factory ecosystem. This may include investing in industrial robotics technologies such as an automated power system to activate equipment when necessary in order to maximize energy efficiency. Sensory and data collection equipment can help businesses find the bottlenecks in manufacturing and address it promptly. Adoption of faster 5g networks allows for real-time improvements. Expansion into the digital ecosystem will allow businesses with an expertise in informational technology to gain a competitive advantage over other businesses that fail to innovate and adapt. Overall, the upgrades in manufacturing will be a major step to better the current production cycle.
Industrialization of ML will become more commonplace in fields such as aerospace, finance, telecommunications, information technology, defense, etc. Companies’ need to understand trends will create a demand for data scientists to manage/collect data and deploy models to fulfill certain goals such as mapping out consumer behavior or detect credit card fraud. The digitalization of ML production helps reliably maintain and scale businesses to unprecedented heights. All in all, it is undeniable that machine learning and artificial intelligence will have a lasting impact in both the landscape of technology and work [4].
In the visual below, McKinsey notes that improvements in machine learning will rapidly accelerate business improvements within the coming years.
MLaaS Market
For trusted product reviews, consumers can turn towards Consumer Reports. However, for ML and AI there is an unregulated market where customers have to make their own decisions: how was this model trained, how accurate is this model, etc. These questions led Stanford data scientist James Zou to develop the dataset HAPI (History of APIs) as a means to compile and review commercially available ML API predictions.
For many of the ML models, they are black boxes, meaning that consumers are usually unaware of how the model will process the inputs into outputs. On top of this, the testing procedure is only known to the developers. To address this, Zou and his colleague Lingjiao Chen created a heterogenous accuracy test that put the models through various datasets to see which products performed better and see how accurate they were over time. HAPI also compared the price of these commercial ML services [5].
In addition to HAPI, there are also efforts from industry organizations and governments to regulate and standardize the MLaaS market. For instance, the Partnership on AI [6], a group of technology companies and academic institutions, has created a set of ethical guidelines for the development and deployment of AI systems. The European Commission has also proposed regulations to ensure that AI systems are trustworthy and respect fundamental rights, including the right to an explanation of automated decision-making processes. These regulations aim to provide a level of transparency and accountability for ML models, giving consumers the assurance that the models have been developed and tested ethically and with due regard for their impact on society.
Smart Workflow
ML and AI revolutionize the way we work. These smart workflows that integrate ML and AI can also take day-to-day work to the next level by automating manual tasks, allowing resources to be allocated elsewhere.
By implementing smart workflows, businesses can not only increase efficiency, but also maximize ROI by 40%-70%. By outsourcing different processes to powerful technology, businesses can expect faster deployment, reduce downtime, and lower operating costs. In terms of marketing/sales, AI-enabled marketing will help organizations focus on real-time feedback from collected data and make better, informed business decisions [7]. One such example is the use of geolocation in smart phones for delivery/gig businesses like Uber and DoorDash. Integrated geolocation technology in the app helps clear up confusion and eases communication between consumers and drivers. In large tech companies, the data collection through smart workflows helps campaign engineers market more effectively.
Generative AI
Generative AI is a type of artificial intelligence that is able to create new data or content, such as text, images, or music, based on a set of learned patterns or rules. There are several types of generative AI, including generative adversarial networks (GANs), variational autoencoders (VAEs), and transformer-based models, each with their own strengths and weaknesses.
One of the most promising applications of generative AI is in the field of machine learning, where it can be used to generate synthetic data for training models, which can be especially useful in cases where real-world data is scarce or expensive to collect. Many businesses today rely on large amounts of data to train their machine learning models, but collecting and labeling this data can be time-consuming and expensive. Generative AI can help alleviate these issues by creating synthetic data that can be used to train models without the need for real-world data. This can be especially useful in fields such as healthcare and finance, where data is often scarce or sensitive.
Another area where generative AI is making an impact is in the field of art and entertainment. Generative AI models have been used to create music, art, and writing, some of which have been indistinguishable from that created by humans. Generative AI can also be used to generate new product ideas, designs, and prototypes, which can save companies time and money by reducing the need for expensive physical testing and prototyping.
As the capabilities of generative AI continue to improve, it is likely that it will play an increasingly important role in a wide range of fields, from healthcare and finance to retail and manufacturing. Additionally, generative AI can be used in automation of repetitive tasks like report generation, customer service and language translation.
In the future, it's likely that generative AI will play an even more important role in business, allowing companies to create new products, services, and experiences that were previously unimaginable. According to Acumen Research and Consulting, the global generative AI market revenue was worth 7.9 billion USD in 2021, with a predicted 34.3% compound annual growth rate (CAGR) from 2022 to 2030 [8].
However, it's important to note that there are also potential downsides to consider, such as the possibility of job displacement and the need for ethical considerations when using generative AI.
Overall, generative AI has the potential to bring significant value to businesses by streamlining processes, reducing costs, and opening up new opportunities for innovation and creativity. It's important for businesses to stay informed about the latest developments in this field and think about how they can incorporate the technology in a way that is both beneficial and ethical.
Blueprint for AI Bill of Rights
In the growing landscape of ML and AI technology, there is an unruly amount of power that these systems deploy. As a means to protect the American people from the abuse of powers, President Biden issued the Blueprint for AI Bills of Rights, a framework to guide the design and deployment of automated systems. The federal government wants to prevent technological threats, which causes discrimination, loss of opportunities, introduces harmful bias, undermines privacy, etc. This framework applies to (1) automated systems that (2) have the potential to meaningfully impact the American public’s rights, opportunities, or access to critical resources or services [9].
This framework describes protections that should be applied with respect to all automated systems that have the potential to meaningfully impact individuals’ or communities’ exercise of: rights, opportunities, or access. The five principles of the framework can be summarized as follows: (1) safe and effective systems, (2) algorithmic discrimination, (3) data privacy, (4) notice and explanation, and (5) human alternative/consideration and fallbacks. The system should be safe and protect its users from outside threats as well as ensure that discrimination is prevented. Users should also be aware of how the system is impacting them and be given autonomy on how their private information is used. They should also be allowed to opt out of technology and ask for human help.
This bill should serve as no surprise to consumers as the federal government is cracking down on data collection from foreign companies. New legislation is putting more scrutiny on China-based apps such as Tiktok, which is currently close to receiving a nationwide ban. The No TikTok on the United States Devices Act pushed by Congress is putting pressure on the White House to ban the social media platform. If this bill is approved, this will serve as a precedent to prevent foreign tech companies from collecting data on US consumers.
Andrew Ng Predicts the Next 10 years in AI
Andrew Ng, founder of LandingAI and DeepLearning.AI, wants to use AI to improve manufacturing which he believes has a “huge impact on everyone’s lives, but is so invisible”. Implementation in warehouses are predicted to hit 57.2% CAGR over the next 5 years [10]. Ng, along with his team at LandingAI, has developed Landing Lens, a platform that helps manufacturing businesses build and deploy visual inspection systems. The challenge was to create a platform to fit each and every manufacturing plant’s needs and the solution was to take a data-centric approach. This approach is focused on keeping the model relatively fixed while fine-tuning the data to train the learning model. Landing Lens gave users the freedom to engineer the data and create their own models using their industry specific knowledge.
In a broader sense, Ng wanted to push the notion that applying domain knowledge using collected data was the best way for machine translation. As models are injected with more and more data, there is less importance to input domain knowledge. However, models with access to limited data are more heavily reliant on domain knowledge to function. This focus on “small data” (data that is comprehensive) is what Ng believes is the next step in advancing AI.
Model Hubs
As AI technology continues to advance and gain wider acceptance, the concept of model hubs has emerged as a potential solution for businesses and organizations looking to harness the power of AI in their operations. Model hubs are centralized platforms that provide access to pre-trained models and other AI tools, making it easier for businesses to adopt and integrate AI into their processes.
One of the main advantages of model hubs is their cost-effectiveness. Model hubs can provide a cost-effective and efficient solution for businesses looking to integrate AI into their operations. This is because model hubs can offer pre-trained models and tools that can be easily adopted and integrated into existing workflows [11]. By providing access to pre-trained models and AI tools, businesses can avoid the high costs associated with developing AI solutions in-house. This can be especially beneficial for smaller organizations or those with limited budgets for AI development.
Another key benefit of model hubs is that they offer access to the latest and most advanced AI technologies and models. This can help businesses to stay competitive in the fast-paced and ever-evolving AI landscape, without having to make significant investments in in-house development. With the rapid pace of technological change in the AI industry, this is a critical factor for businesses looking to stay ahead of the curve.
Model hubs also foster collaboration and sharing, allowing organizations to share and reuse models and tools. This helps to reduce the need for duplication of effort and increase the efficiency of AI resource usage. It also helps to ensure that models are developed to a high standard, as they are tested and validated by a wider community of users.
In addition, model hubs provide a centralized and accessible platform for AI development and deployment, making it easier for businesses to adopt and integrate AI into their processes. This can help organizations to increase their operational efficiency and productivity, while also unlocking new business opportunities and capabilities.
Overall, the future of model hubs in the AI industry looks bright, as they offer a cost-effective and efficient solution for businesses looking to integrate AI into their operations. With the growing demand for AI technology, model hubs are set to play a key role in helping organizations to stay competitive and keep pace with the fast-moving AI landscape.
Conclusions
[Ryan] In conclusion, the future of AI and technology is looking bright, with numerous advancements and innovations happening in the field. AI is poised to play a crucial role in various aspects of society, from work and education to healthcare and entertainment. The development of cutting-edge AI and ML technologies has the potential to revolutionize various industries and bring about significant benefits for individuals and businesses alike. As technology continues to advance at an unprecedented pace, the opportunities for AI to have an even greater impact on our world are endless. It is up to us to harness its potential in ethical and responsible ways, to ensure a brighter future for all.
Moreover, as AI becomes increasingly integrated into our lives, the ethical implications of its use become even more important. It is crucial that we consider the ethical and societal implications of AI and work to ensure its development and implementation aligns with human values and ethics. This includes the responsible use of data, protection of privacy, fair and unbiased decision-making, and the reduction of AI's potential negative impact on society. The future of AI must prioritize transparency, accountability, and ethical considerations to ensure its responsible and beneficial integration into our lives. Addressing these ethical concerns and shaping the future of AI in a responsible manner will be a critical challenge in the years to come.
[Duy] With the surge of new ML and AI technology, I believe that the future of work is being revolutionized to new heights and AI will be indispensable to every corporation. Menial tasks like data collecting, accounting, maintaining records, etc. will be outsourced to automated systems and human labor will be allocated to more complex tasks. Blue collar jobs such as truck driving may see a decline with companies like Waymo, Tesla, and Volvo developing self-autonomous driving. The application of ML techniques has lowered the barrier to conduct science experiments and will propel the research in chemistry, pharmacology, agronomy, etc. ML is expected to integrate into all aspects of life and provide businesses with demonstrable returns on investments.
ML technology is already making waves in all sectors whether it be transportation or law. Over the years, previous models have had 65% accuracy, but now the standard BERT-based models can produce accuracy of 85%. Higher accuracy allows LLM (language learning models) to be more versatile rather than just classification; an example is GitHub Copilot which is a cloud-based AI tool used to develop code. LLM can also be used to help ecommerce businesses create product descriptions and enhance customer services/support. With so many automation tools at one’s disposal, ML technologies will inevitably improve work efficiency. With the rise of ChatGPT, the newest chatbot developed by OpenAI, Google’s entire search engine business model is threatened; an LLM model gives direct answers instead of endless pages of ads. Regardless of its temporary bugs and faults, the future for ChatGPT and LLM is promising. What’s truly stopping artificial intelligence from becoming a consultant for finance, business, or even medicine?
The passing of new regulations does pose a concern to emerging ML technology. Only time will tell how new legislation will affect the future of artificial intelligence.
About Us
Loxz Digital Group is a Machine Learning Collective located in Berkeley, CA. Established in December of 2020, Loxz is focused on serving RealTime predictive analytics. We supply models and serve predictions within smart workflows to clients of the Email Service Provider network and enterprises and are in discussions with serving location specific predictions to law enforcement to reduce gunshot violence. We employ a servant-leadership management style where every employee or advisor has a distinct voice. Specifically, RealtimeML is at the bedrock of what we do. Collectively, the current assembled team has over 40 years of ML experience, housing 8 data scientists, all located in the United States and Canada.
©2023 All Rights Reserved.
Visit www.loxz.com
References
- [1] Ransbotham, S., Kiron, D., Gerbert, P., & Reeves, M. (2017). Reshaping business with artificial intelligence: Closing the gap between ambition and action. MIT Sloan Management Review, p. 59.
- [2] Proficient Market Insights. “Artificial Intelligence (AI) Market Projected to Grow at a Magnificent Cagr during the 2022-2028 Forecast Timeframe.” GlobeNewswire News Room, Proficient Market Insights, 10 Nov. 2022, arket-Projected-to-Grow-at-a-magnificent-CAGR-During-the-2022-2028-Forecast-Timeframe-98-Pages- Report.html .
- [3] “Smart Factory Market (by Product: Machine Vision Systems, Industrial Robotics, Control Devices, Sensors, Communication Technologies, Other Products; by Technology: Product Lifecycle Management, Human Machine Interface, Enterprise Resource and Planning, Distributed Control System, Manufacturing Execution System, Programmable Logic Controller, Supervisory Controller and Data Acquisition; by End-User Industry) - Global Industry Analysis, Size, Share, Growth, Trends, Regional Outlook, and Forecast 2023-2032.” Precedence Research, www.precedenceresearch.com/smart-factory-market.
-
[4] “2022: The Year in Charts.” McKinsey & Company, 9 Dec. 2022,
- [5] Miller, Katharine. “A Consumer Reports for AI Services.” Stanford HAI, 9 Jan. 2023, hai.stanford.edu/news/consumer-reports-ai-services?utm_source=Stanford%2BHAI&utm_campaign= ed0133cbe7-Mailchimp_HAI_Newsletter_January%2B2023_2&utm_medium=email&utm_term=0_aaf0 4f4a4b-25ccfb53b4-%5BLIST_EMAIL_ID%5D.
- [6] Partnership on AI. (2023, January 18). Retrieved January 30, 2023, from https://partnershiponai.org/
- [7] Gillis, Alexander S. “What Is Intelligent Workflow and What Are Its Benefits?” WhatIs.com, TechTarget, 21 Apr. 2021, www.techtarget.com/whatis/definition/intelligent-workflow#:~:text=Intelligent%20workflow%20is%20 an%20orchestration,necessary%20to%20complete%20a%20task
- [8] Acumen Research and Consulting. (2022, December 14). Generative AI market size will achieve USD 110.8 billion by 2030 growing at 34.3% CAGR - exclusive report by Acumen Research and consulting. GlobeNewswire News Room. Retrieved January 26, 2023, from https://www.globenewswire.com/news-release/2022/12/14/2574140/0/en/Generative-AI-Market-Siz e-Will-Achieve-USD-110-8-Billion-by-2030-growing-at-34-3-CAGR-Exclusive-Report-by-Acumen-Researc h-and-Consulting.html#:~:text=TOKYO%2C%20Dec.,34.3%25%20from%202022%20to%202030.
- [9] “Blueprint for an AI Bill of Rights.” The White House, The United States Government, 4 Oct. 2022, www.whitehouse.gov/ostp/ai-bill-of-rights/#safe.
- [10] Anadiotis, George. “Andrew Ng Predicts the next 10 Years in Ai.” VentureBeat, VentureBeat, 21 Mar. 2022, venturebeat.com/ai/andrew-ng-predicts-the-next-10-years-in-ai/?utm_campaign=c891300a87-EMAIL_ CAMPAIGN_6_28_2022_11_1_COPY_01&utm_medium=email&utm_source=Towards%2BAI&utm_term =0_324cc77d4c-c891300a87-599552348
- [11] Robinson, B. (2022, November 9). 'Hub-and-Spoke': The new office model of the future, expert says. Forbes. Retrieved January 30, 2023, from https://www.forbes.com/sites/bryanrobinson/2021/06/09/hub-and-spoke-the-new-office-model-of-th e-future-expert-says/?sh=35bf62022732
-
Figure 1
“Smart Factory Market (by Product: Machine Vision Systems, Industrial Robotics, Control Devices, Sensors, Communication Technologies, Other Products; by Technology: Product Lifecycle Management, Human Machine Interface, Enterprise Resource and Planning, Distributed Control System, Manufacturing Execution System, Programmable Logic Controller, Supervisory Controller and Data Acquisition; by End-User Industry) - Global Industry Analysis, Size, Share, Growth, Trends, Regional Outlook, and Forecast 2023-2032.” Precedence Research, https://www.precedenceresearch.com/smart-factory-market. -
Figure 2
Chui, Michael, et al. “McKinsey Technology Trends Outlook 2022.” McKinsey & Company, McKinsey & Company, 31 Oct. 2022, https://www.mckinsey.com/capabilities/mckinsey-digital/our-insights/the-top-trends-in-tech?cid=othe r-eml-alt-mip-mck&hlkid=fada852ff1084e38ae7152f763e6e735&hctky=9972939&hdpid=41a88839-08 38-4428-b64f-620ab8d987c9. -
Figure 3
Chui, Michael, et al. “McKinsey Technology Trends Outlook 2022.” McKinsey & Company, McKinsey & Company, 31 Oct. 2022, https://www.mckinsey.com/capabilities/mckinsey-digital/our-insights/the-top-trends-in-tech?cid=othe r-eml-alt-mip-mck&hlkid=fada852ff1084e38ae7152f763e6e735&hctky=9972939&hdpid=41a88839-08 38-4428-b64f-620ab8d987c9. -
Figure 4
Acumen Research and Consulting. (2022, December 14). Generative AI market size will achieve USD 110.8 billion by 2030 growing at 34.3% CAGR - exclusive report by Acumen Research and consulting. GlobeNewswire News Room. Retrieved January 26, 2023, from https://www.globenewswire.com/news-release/2022/12/14/2574140/0/en/Generative-AI-Market-Siz e-Will-Achieve-USD-110-8-Billion-by-2030-growing-at-34-3-CAGR-Exclusive-Report-by-Acumen-Researc h-and-Consulting.html#:~:text=TOKYO%2C%20Dec.,34.3%25%20from%202022%20to%202030 -
Figure 5
Chui, Michael, et al. “McKinsey Technology Trends Outlook 2022.” McKinsey & Company, McKinsey & Company, 31 Oct. 2022, https://www.mckinsey.com/capabilities/mckinsey-digital/our-insights/the-top-trends-in-tech?cid=othe r-eml-alt-mip-mck&hlkid=fada852ff1084e38ae7152f763e6e735&hctky=9972939&hdpid=41a88839-08 38-4428-b64f-620ab8d987c9 -
Figure 6
Bibliography – Digital Humanities. digital-humanities.info/bibliography
Contributors
Ryan Peng, Data Scientist Lead,
Lead Author, Lead Analyst
Duy-Anh dang, Data Scientist Intern,
Report Designer
Yumi Koyanagi, Designer,
Report Designer
