Why We’ll Be Watching These 5 AI Trends in 2024 Very Closely
As a trusted AI governance software provider, we’re not just keeping an eye on AI trends in 2024 – we’re glued to the screen.
You can think of us as your AI watchdog, and it’s our job to help organizations navigate the AI maze safely and securely.
As we step into 2024, we’re on the cusp of many groundbreaking advancements in artificial intelligence. Whether you’re an industry expert or an intrigued observer, these trends offer a glimpse into a future where technology and innovation intersect in absolutely fascinating ways.
Welcome to the Future of AI.
#1. LLM Extensions Boom
2024 will be a pivotal year for foundational LLM innovation and extensions in multimodal (text, images, voice, video) and search.
We will be watching Google’s Gemini and JPMorgan’s DocLLM very closely this year. Both are examples of significant innovations in multimodal AI. Gemini stands out for its specialization in language and dialogue understanding, while DocLLM represents a significant leap in the intricate processing and comprehension of legal and financial documents. Both these advancements signal a transformative phase in the application of artificial intelligence across a wide range of domains.
Meanwhile, perplexity.ai has just raised $250M series B in its attempt to disrupt Google with an intuitive, conversational search experience. Retrieval-augmented generation (RAG) techniques allow LLMs to incorporate corpora of new information without expensive retraining or fine-tuning, meaning models will always have access to the latest information – a must for effective search capabilities.
Expect continued innovation in foundational models, especially in the application layer, and custom generative AI technologies that are domain- or function-specific and sit on top of foundational models.
Want to learn more about this space? Follow
#2. Regulations, Standards, & Lawsuits May Provide Short-Term Confusion (And Long-Term Clarity)
The past year was just the start of a long AI regulation and standards journey. We saw the enforcement of NYC LL144 in July, which requires bias audits of AI used in hiring. The EU AI Act passed in December, will set AI requirements across all EU member states. Many other bills were introduced and debated globally.
Several critical frameworks and standards were introduced, most notably the NIST AI Risk Management Framework and ISO/IEC 42001.
In 2024, the momentum will only continue. The final text of the EU AI Act will be released early this year, and companies will know the requirements and have time to get compliant before enforcement starts in 2025.
With the introduction of ISO/IEC 42001, companies now have an assessment-based way to demonstrate the soundness of their AI management practices. Adopting and proving conformity with responsible AI standards will become an expectation, much like SOC2 and ISO/IEC 27001 are today for data security.
In addition to regulations and standards, landmark lawsuits will also help provide clarity. For example, the adjudication of the NYTimes’s lawsuit against OpenAI will go a long way to clarifying principles around data usage, ownership, and copyright infringement for AI training.
Want to learn more about AI regulation? Follow these experts for updates on the most notable advances in policy (and the lawsuits that are sure to follow):
- Reva Schwartz, Research Scientist/Principal Investigator for AI Bias at the National Institute of Standards and Technology (NIST)
- Keith Sonderling, Commissioner at the United States Equal Employment Opportunity Commission (EEOC)
#3. Significant growth of AI Governance Software and Corporate AI Trust, Risk, and Security Management (AI TRiSM) programs.
While we may look back on 2023 as the “wild, wild, west” of AI, 2024 is set to usher in much-needed guardrails and guidelines for AI use in the workplace.
First coined by Gartner in 2023, AI TRiSM refers to the burgeoning field of AI governance software and tools aimed at creating accountability and risk management processes within the workplace.
This suite of software looks to address the many challenges of AI systems in the workplace including fairness, governance, explainability, reliability, and privacy.
“AI is the future, and companies that don’t use AI will be at a severe competitive disadvantage. But there are also many, many ways it can go wrong, so being well-managed and having AI governance is important.“ Guru Sethupathy, CEO & Co-Founder of FairNow, How to Ensure AI in HR is Fair, Effective and Explainable
Expect to see many conversations around accountability and risk management as wise companies recognize that using AI to its full potential requires a combination of both enthusiasm and caution. Finding the right balance between the two is the key to leveraging AI effectively and responsibly.
Follow these industry experts for the latest advances in AI Governance technology:
- Avivah Litan, Distinguished VP Analyst, Gartner
- Guru Sethupathy, CEO & Co-Founder of FairNow
- Christina Montgomery, Head of AI Ethics, Chief Privacy & Trust Officer at IBM
#4. Rapid Exploration and Adoption of GenAI In The Corporate World
Large companies will significantly expand their exploration of generative AI in 2024. Budget space was created at the end of the year (to the tune of $500B, according to some estimates) to pilot, experiment, and adopt generative AI.
Multiple trends are converging to create the perfect storm to drive adoption in 2024. The first is more legal certainty and clearer guardrails around generative AI usage, which was cited as a top challenge for generative AI adoption in 2023. The second trend is an increased maturity of tools. Offerings from the major cloud providers and other companies make it easier to access and experiment with generative AI tools in a way that wasn’t possible last year. And finally, as companies experiment, explore, and iterate, they will find more and better use cases for generative AI.
Follow these business leaders as they continue to bring responsible AI to the corporate world:
- Prem Natarajan, Chief Scientist & Head Of Enterprise AI at Capital One
- Rajamma Krishnamurthy, Senior Director Of Enterprise AI Strategy at Microsoft
#5. Open-Source Models Close The Gap In Power And Capabilities
Finally, open source will play a more prominent role in foundational models and related ecosystems. Part of the reason is that foundational models may not have sufficient moats, but rather, the value is created through the ecosystem. Second, open-source models (e.g., Mistral) are already very competitive.
While closed-source models were the most capable models in 2023, open-source models are closing the gap in both power and capabilities. Open-source foundational models also benefit from a broader pool of contributors to the ecosystem, increased transparency, and more democratized guardrails.
Follow these thought leaders to understand why many industry experts have expressed the importance of open-source models in 2024:
- Yann Lecunn, Chief AI Scientist at Meta, Facebook AI Research (FAIR)
- Guillaume Lample, Co-Founder & Chief Scientist at Mistral
Now We Wait
And there you have it – the five AI trends that are setting the tone for 2024.
This year promises to be a remarkable period for AI, marked by significant advancements and critical developments in governance and ethical standards. By understanding and embracing these trends, we can harness the full potential of AI to drive innovation, enhance efficiency, and create a more intelligent and connected future. See you there!