Just over a year after an information overload (almost always exaggerated and misinformed) surrounding Artificial Intelligence, or AI, it's time to move from conversation to implementation. Countless opportunities already exist to train in basic skills for free, so now is the time to put them into practice.
If you're not yet familiar with the fundamentals of AI, simply browse Coursera, edX, or Platzi to find free courses that will help you understand the general context. There are even specialized curricula for specific areas of knowledge, such as our "Introduction to Artificial Intelligence for Librarians," which remains available, free, and open to everyone. www.iaparabibliotecas.com (with an initial session suitable for any profession and academic level).
In this article, we'll take a quick look at AI in 2023, along with what 2024 has in store, to help us propose some key dynamics for its implementation. Welcome to AI: This is what intelligence looks like!
2023: Awakening from the shadows of Generative AI
2023 was pivotal for Artificial Intelligence (AI), marking a year of significant transformation and rapid advancements. The integration of Generative AI (GenAI) into various business functions signals a shift towards More innovative and efficient operations, although with emerging challenges related to accuracy, cybersecurity and regulatory compliance. Leading organizations (OpenAI/Microsoft, Anthropic/Amazon, Google, Meta, ElevenLabs, Runway ML, Stability, Perplexity, etc.) are leveraging AI not only to help reduce costs, but also as a fundamental component of product and service development, emphasizing the creation of new businesses and enhancing the value of existing offerings. These entities are investing significantly in AI, integrating it into multiple business functions, demonstrating the strategic importance of AI in driving competitive advantage and operational excellence. (https://www.mckinsey.com/capabilities/quantumblack/our-insights/the-state-of-ai-in-2023-generative-AIs-breakout-year).
The GPT-4 presentation underscored the paramount role of computing power, comparing it to the new oil fueling the rapid evolution of AI. However, There is growing concern about the energy consumption and carbon emissions resulting from training massive models (https://www.stateof.ai/).
He Hype Cycle Gartner's 2023 AI Outlook identifies generative AI and decision intelligence as innovations poised for widespread adoption in the coming years. This reflects AI's potential to revolutionize businesses if deployed responsibly and effectively. (https://www.gartner.com/en/articles/what-s-new-in-artificial-intelligence-from-the-2023-gartner-hype-cycle).
From accelerating scientific discoveries to raising ethical dilemmas, AI shows its double edge. AI has become a powerful tool for advancement and a subject of debate regarding its implications for society. With AI models contributing to major scientific and technological breakthroughs, the discourse surrounding impartiality, bias, and ethics is gaining prominence, reflecting the complex interplay between technological innovation and social norms. (https://aiindex.stanford.edu/report/).
2024: Navigating Uncertainty in the Age of AI
In recent weeks we have seen the emergence of tools that continue to challenge our ability to discern between the real and the digitally generated, such as OpenAI's Sora video creator (https://openai.com/soraUncertainty, like change before it, is a new constant for human societies in 2024: what do we know will come? What has uncertainty already brought us?
Is it the realization of what marketing promised?
The hype surrounding AI, which reached new heights in 2023, is now facing a reality check. Companies are expected to weigh the actual benefits of AI tools against their costs and risks, which could lead to a Retraction of premature or excessive AI implementations. This reassessment could also trigger legal and regulatory action against AI service providers who fail to deliver on their promises. (https://techcrunch.com/2023/12/19/8-predictions-for-ai-in-2024/).
Apple's entry into AI: SiriX
Apple is about to make a major foray into AI, a nnew generation of its iPhone focused on improving personal digital assistant (Siri) through data and at the same time avoiding the generation of texts, images or media that are risky for privacy.
A Generative AI trained with your own data
Generative AI (GenAI) will extend its influence beyond creative applications, becoming an integral part of content strategy and product or service design. Improvements in Recall Augmented Generation (RAG, a technique for using proprietary data in AI “retraining”) will enable to obtain more accurate and contextually nuanced results, using large amounts of own information to tune or retrain AI models (https://www.veritas.com/blogs/ai-in-2024-10-predictions-shaping-our-future). We have already seen the launch of a paid version of Gemini, Google's foundational AI, which promises to learn and adapt to the content in our Google Drive (https://gemini.google.com/advanced).
AI-generated disinformation
AI-generated disinformation, especially in the context of elections, is poised to become a major challenge. It is expected that...The ease of creating deepfakes and the proliferation of disinformation complicate the outlook political and the information ecosystem in general (https://www.technologyreview.com/2024/01/04/1086046/whats-next-for-ai-in-2024/We have already seen some joint government and industry initiatives to combat disinformation in the context of the 2024 elections, such as the AI Elections Accord. https://www.aielectionsaccord.com/.
Trade and innovations in security
In retail, virtual shopping advisors (powered by GenAI) will offer personalized shopping experiences, while AI-driven security systems will enhance in-store safety. This dual application underscores AI's role in optimizing customer engagement and operational security. (https://blogs.nvidia.com/blog/2024-ai-predictions/).
Industrial digitization and AI
The fusion of industrial digitization with GenAI is poised to accelerate industrial transformation, particularly in the field of robotics, by 2024. This will facilitate the conversion of physical attributes into digital data in both directions, thereby improving product design, manufacturing, virtual training, and worker safety. We are already seeing initiatives where LLMs (degrees in language modeling) are integrated with robots to explore possibilities in various industries (https://www.scientificamerican.com/article/scientists-are-putting-chatgpt-brains-inside-robot-bodies-what-could-possibly-go-wrong/)
Legal and compliance overview
The legal and regulatory landscape surrounding AI is expected to become a battleground, with debates on fair use, copyright, and civil liability heating up alongside efforts to regulate major linguistic models and specific use cases (https://www2.deloitte.com/us/en/pages/technology/articles/ai-predictions-for-2024.htmlThis represents an opportunity for civil society to enter the conversation and enrich the debate surrounding human rights and technology (already addressed by regional impact organizations such as the Karisma Foundation in https://web.karisma.org.co/lanzamos-mi-pequeno-glosario-de-derechos-humanos-y-tecnologia/)
Challenges and innovations in the AI engine: chips
A potential global shortage of GPUs (Graphics Processors), crucial for AI operations, will accelerate innovation in hardware development, driving efforts to find low-power and cost-effective alternatives to current GPUs. (https://hai.stanford.edu/news/what-expect-ai-2024Google, Meta (Facebook), and Intel are already in advanced stages of developing GPUs to rival the undisputed market leaders: Nvidia's H100. Innovative initiatives like Groq, which have completely redesigned processors to accelerate interaction with chatbots by up to 1000x, They are beginning to emerge in the market and we can already see their results (https://groq.com/).
Myths and realities: no apocalypse in sight
Contrary to catastrophic predictions, AI is not expected to eliminate all jobs or end civilization.. On the contrary, it will continue to help organizations that adopt it responsibly to make better decisions, highlighting AI's role as a tool for improvement, not displacement. (https://www.sas.com/en_us/news/press-releases/2023/november/ai-predictions-2024.html).
These predictions for 2024 underscore the dynamic nature of AI development and its growing impact across various sectors. As AI continues to evolve, businesses, policymakers, and individuals alike must navigate these changes with a critical and informed approach, balancing innovation with ethical considerations and the well-being of society. So how do we move from talking about it to implementing Artificial Intelligence in 2024?
Principles for a conscious implementation of AI in 2024
The conscious use of Artificial Intelligence (AI) involves Integrate ethical guidelines, equity, inclusivity, and safety into the development and deployment of AI technologies. This approach is vital in all sectors, including business, education, culture, healthcare, and technological development.
Ethical design and equity
● Commitment to diverse experts: It is increasingly crucial to involve social and human scientists, ethicists, and other relevant experts in the AI development process to ensure that diverse perspectives are considered. This helps to design AI models that are fair and inclusive, taking into account the impacts on different user groups and communities. (https://ai.google/responsibility/responsible-ai-practices/).
● Diverse and representative datasets: Applying infodiversity principles to the training, tuning, and testing of AI models is essential to avoid bias. Fairness in AI requires continuous evaluation of data representation and the mitigation of discriminatory biases.
Interdisciplinary collaboration in implementation
The implementation of AI should be considered a challenge that is as much human as technological, requiring collaborative ecosystems that include AI developers, social and human scientists, user experience (UX) designers, lawyers, ethicists, business leaders, and users. This collaborative approach ensures a responsible deployment of AI technologies. (https://www.bcg.com/capabilities/artificial-intelligence/ai-for-business-society-individuals/responsible-ai), where key questions are raised:
- What are the processes that require the most cognitive effort in the organization?.
- Among these processes, which ones have a lower emotional return for the team (satisfaction in their execution)?.
- Which Artificial Intelligences can help to semi-automate or automate those processes?.
- What are the ethical, legal, and reputational implications of using these AIs?.
- Which soft skills should I develop in the team before using AI?.
- How to train the team in the use of those AIs.
- How will the success of that implementation be measured and how will continuous learning be promoted?.
Creating a framework for responsible use
A simple yet effective framework for the responsible use of AI includes evaluating initial results, verifying facts and data, editing indications to improve them, and reviewing results to ensure accuracy and transparency. Ultimately, lThe responsibility for ensuring that AI results meet ethical standards lies with the end users., organizations or people (https://www.aiforeducation.io/ai-resources/how-to-use-ai-responsibly-every-time).
Management of new organizational risks
With the advancement of generative AI, new risks emerge, including those related to cybersecurity, privacy, legal issues, performance, bias, and intellectual property. To mitigate these risks, it is essential that senior executives have a thorough understanding of responsible AI practices. (https://www.pwc.com/us/en/tech-effect/ai-analytics/responsible-ai-for-generative-ai.html). Are we preparing managers while training the team for implementation?
Specific considerations for the health sector
In healthcare, the deployment of AI requires careful attention to biases, which can lead to unequal treatment and detrimental outcomes for patients. Transparency in how patient data is used, along with strict security measures, is essential to protect sensitive information and ensure patient health and privacy. (https://healthitanalytics.com/features/responsible-ai-deployment-in-healthcare-requires-collaboration).
By adhering to these principles and practices, organizations canharnessing the power of AI in ways that benefit society while mitigating potential harms. The conscious use of AI is not just about avoiding risks, but also about harnessing it as a force for good, promoting equity and inclusion and improving lives.
And what steps have you taken for a conscious implementation of AI in your personal and professional life?
English 
Español