Artificial Intelligence (AI) is not a new concept; it’s been around since the early 1950s, but limitations in computing power and algorithms have restricted this market’s growth. Establishing and operating an AI infrastructure was prohibitively expensive, qualified staff scarce, and scalability limited, narrowing access to large organizations such as research institutions, big technology and financial services firms, and public sector. Not anymore. In recent years, especially during and after the pandemic, there’s been a significant advancement in the use of AI, boosted by cloud computing, which provides accessibility, usability, and scalability.
Up to now, chatbots – software designed to simulate a human conversation – have been the most prevalent use case; these were introduced by many companies to deal with the increased demand for customer service in digital channels. Most chatbots still rely on pre-configured rules and responses. However, more recently, large language models (LLMs) are being trained to understand the questions of customers and respond in a natural way, like the ideal human agent, but much faster. Such models are constantly learning based on the interactions and feedback, thus improving accuracy and usefulness of responses.
These advanced chatbots make use of what the market is now referring to as Generative AI (GenAI), a subgroup of AI that enables users to generate text, images, videos, presentations, and other outputs based on text prompts or voice interaction with LLMs. In the past couple of years, groundbreaking advancements in commercial LLMs have come from technology companies such as AI21 Labs, Amazon, Anthropic, Cohere, Google, OpenAI, NVIDIA, Meta and Microsoft. Software firms have been embedding GenAI in their consumer and enterprise applications. Beyond GenIA, as other forms of AI extend their reach further, demand for applications such as facial recognition, virtual assistants, productivity enhancers, and recommendation algorithms is growing -and fast.
Yet, the AI-driven revolution poses a challenge for emerging economies in which enterprises still lack technology maturity and are more subject to restrictive purchasing power and uncertain political and economic scenarios. In Latin American countries, most companies are not prepared to implement the data frameworks that AI relies on, such as data governance, data quality, data security, data architecture, data management platforms and data operations. Lack of AI-proficient staff, budget constraints, and unpredictable political or economic factors join forces to curb AI implementation.
But even under challenging circumstances, businesses’ awareness regarding the urgent need to adopt AI technology is growing. Many are feeling the threat of lagging behind competitors who manage to adopt new technologies faster, making a leap forward in productivity and customer experience that might be unmatched in a few years—or even months. In Latin America, Frost & Sullivan’s latest decision-maker survey, conducted in Q4 2023, anticipates this movement: the implementation of Artificial Intelligence and Machine Learning is already the top technological investment priority in 2024, surpassing cloud computing, which in many cases is the supporting technology platform. Eighty-nine percent of companies in the region believe that AI will be “disruptive” to some extent. Nevertheless, the same data indicated that only 29% of companies in Latin America are building or planning to build GenAI solutions in-house in the next year, while 71% are expecting the market to provide them with AI solutions.
For those countless companies with AI as a priority but struggling to formulate a concrete initiative, cloud providers are already offering valuable solutions. Amazon Web Services (AWS), the leading cloud computing service provider in Latin America, has significantly invested in services based on AI and is ready to support the needs of enterprise users requiring the use of artificial intelligence in their apps and operations. The cornerstone of this portfolio evolution is Amazon Bedrock. The solution was launched in September 2023 in selected data regions in the world. In Latin America, by now tens of thousands of companies are currently utilizing Amazon Bedrock hosted in the North American data regions of AWS.
The excellent news is that since June 7, 2024, Latin American enterprises can access the Amazon Bedrock service from data centers located in the Sao Paulo-Brazil region. With this newly available spot, Latin American firms can deploy AI workloads with minimal latency and adhere to regulations such as data localization and sector-specific regulations such as those governing financial data or healthcare data.
But what value does Amazon Bedrock bring to enterprises in Latin America?
First, it is a managed service designed to streamline the creation and scaling of applications based on GenAI models. It allows developers to access a variety of third-party and AWS-owned foundational models to build applications capable of generating text, images, audio, and other types of content. Bedrock offers enterprise users a selection of high-performance foundation models (FMs) from leading AI companies such as AI21 Labs, Anthropic, Cohere, Meta, Mistral AI, Stability AI, and Amazon, all through a single API.
Second, it allows companies to test and evaluate the most advanced foundation models for their specific needs, and then customize them privately with their own data using techniques like fine-tuning and retrieval-augmented generation (RAG). Users can also create agents to perform tasks using their own enterprise systems and data sources. Bedrock operates as a serverless service, removing the need for infrastructure management and facilitating the secure integration and deployment of generative AI capabilities into applications through AWS services. As an addition, AWS has also launched Generative AI Application Builder (for companies lacking specific know-how to develop AI apps) and Amazon Q (generative AI-powered assistant for accelerating software development and leveraging companies’ internal data).
An important customer of AWS in the region using Amazon Bedrock is iFood, the leading Brazilian platform for online food ordering and delivery. iFood used models such as Anthropic’s Claude, Amazon Titan, and retrieval-augmented generation on Amazon Bedrock to build a proof of concept called “Garcon”, a virtual waiter that interacts with users and takes voice/text orders using GenAI.
Other interesting AWS cases of GenAI in the region are Gimba, an e-commerce for office and IT supplies in Brazil, which is using GenAI to automate stages of building its product catalog of approximately 30 thousand items, reducing product registration time by 84%; and Indra Energia, a Brazilian energy company that adopted GenAI with Amazon Bedrock in its platform to allow the migration of millions of users to the free energy market with tailored AI-generated solutions.
To provide proper support to AI advancements, AWS has invested significantly in recent years to equip its infrastructure with sufficient means to sustain AI resource-intensive workloads at a low cost, at a time when optimization has become one of the foremost concerns for cloud-operating businesses. The first significant step in this direction was the custom-processor Graviton, a 64-bit ARM-based general-purpose processor designed to deliver enhanced performance for general cloud workloads at a reduced cost (over 2 million of them are already deployed worldwide). It was followed by AWS Inferentia, the first custom-designed accelerator aimed at providing accelerated real-time inference with machine learning and generative AI models. Finally, in 2020, AWS introduced Trainium, a specialized accelerator chip developed for pre-training and fine-tuning large-scale generative AI and deep learning models.
Shaking a leg
The world is still in the early days of the artificial intelligence era. For a newcomer, it may still seem that only large companies in the technology sector will benefit from this new industrial revolution. Up to this moment, for most users, AI has largely meant a chatbot that can answer to almost anything. However, the transformation is much deeper within the business world. There are good reasons (and facts) to argue that large-scale enterprise adoption of AI, across all industries, has begun. As the experimental phase comes to an end, companies worldwide are planning to implement AI.
Under this large-scale tectonic movement, the operations of the companies whose services we use daily will begin to transform radically; they will gain efficiency and speed, and in addition they will be able to offer better quality products and services or previously nonexistent, and even unimaginable just a short time ago. We can’t overlook the fact that this will pose a relevant challenge for businesses. The key lies in finding ways to organically integrate AI, with tools that simplify a process that, at first glance, may seem insurmountable to many. But, although AI may seem taken out of science fiction, with the new enterprise services available, ensuring its safe adoption and operation can be a much more worldly task.
About Frost & Sullivan:
Frost & Sullivan, the growth pipeline company, enables clients to accelerate growth and achieve best-in-class positions in growth, innovation, and leadership. The company’s Growth Pipeline as a Service provides the CEO’s Growth Team with transformational strategies and best-practice models to drive the generation, evaluation, and implementation of powerful growth opportunities. Let us coach you on your transformational journey, while we actively support you in fostering collaborative initiatives within your industry’s ecosystem. Our transformation journey is fueled by four powerful components, ensuring your success in navigating the ever-changing landscape of your industry.
- Schedule a complementary Growth Dialog with our team to dive deeper into transformational strategies and explore specific needs within your company.
- Become a Frost Growth Expert in your area of specialization and share your expertise and passion with the community through our think tanks.
- Join Frost & Sullivan’s Growth Council and become an integral member of a dynamic community focused on identifying growth opportunities and addressing critical challenges that influence your industry.
- Designate your company as a Companies2Action to increase exposure to investors, new M&A opportunities, and other growth prospects for your business.
