As we leap into 2025, AI is sprinting ahead, reshaping industries, careers, and how we live. Back in 2024, Srikanth Velamakanni, Fractal.ai’s co-founder, made bold AI predictions. Did they hit the mark? Let’s find out!
Srikanth Velamakanni’s 2024 Predictions
The first five predictions focused on Large Language Models (LLMs) and Foundation Models. Here’s what he envisioned and where we stand today.
The Foundation Model Race will Heat-up
The competition to create the most powerful AI models is accelerating. Gemini, a new AI model, was expected to launch before February. However, the prediction that GPT-5 would surpass Gemini by March turned out to be partially true.
Verdict: Partaily True 🔄
Open-Source AI Models will Grow
Open-source AI models are improving rapidly and are now close to the capabilities of GPT-4. This prediction turned out to be true, as advancements in open-source models have made them more accessible and powerful.
Verdict: True ✅
Reduced Token Costs
The cost of using AI tools has significantly dropped, with a reduction of over 75% compared to 2023. This prediction was also true, making AI more affordable for businesses and developers.
Verdict: True ✅
Foundation Models vs. Domain-Specific Models
General-purpose AI models (foundation models) are still competing with specialized models designed for specific tasks. However, hybrid approaches that combine both have proven to be valuable. This prediction is true as both approaches continue to coexist and complement each other.
Verdict: Partaily True 🔄
Rise of AI Agents
Customizable AI agents, which simplify task automation, decision-making, and information retrieval, have gained popularity. This prediction is true, as these agents are increasingly used across various industries to streamline workflows.
Verdict: True ✅
The next five predictions focus on microeconomics and startups.
Productivity Boost Across Sectors
AI assistants, applications, and agents are expected to significantly increase productivity by handling repetitive and creative tasks. This prediction turned out to be true as AI is helping industries work faster and more efficiently.
Verdict: True ✅
Jobless Growth in Key Sectors
While industries are growing due to AI advancements, this growth is not necessarily creating more jobs. The prediction that many sectors would see jobless growth is partially true, as some industries are experiencing this trend while others are not.
Verdict: Partaily True 🔄
Reduced Funding Sizes in Early Stages
It was predicted that early-stage startups would require less funding because AI makes it cheaper to build and launch products. However, this turned out to be false, as funding requirements remain significant in many cases.
Verdict: False ❌
Costlier Horizontal AI Companies
Companies that aim to build broad-scope AI solutions, like DeepMind or OpenAI, are becoming more expensive to run due to rising talent and training costs. This prediction is true as the resources needed to create such advanced systems continue to grow.
Verdict: True ✅
AI Agents Gone Rogue
The expectation of widespread news about AI mishaps, ranging from funny errors to serious issues, has been particially true. While there have been occasional mishaps, they haven’t been as frequent or severe as expected.
Verdict: Partaily True 🔄
The next five predictions focus on enterprise AI, showcasing its increasing role in transforming business functions. While some advancements have surpassed expectations, others are still a work in progress.
Enterprise Adoption of Generative AI
Over 80% of top companies, like those in the Fortune 100, are expected to start using Generative AI for various tasks as privacy and compliance risks are addressed. This prediction is true as many enterprises have already embraced Generative AI in their operations.
Verdict: True ✅
Enhanced Enterprise and Website Search
Generative AI is expected to make searching within companies and on websites much more efficient and effective. This prediction is true as AI-powered search tools are becoming significantly better.
Verdict: True ✅
Enterprise BI + AI
The combination of Business Intelligence (BI) and Artificial Intelligence (AI) will improve predictions, insights, and recommendations for businesses. This prediction is also true as more enterprises are leveraging this powerful duo to make data-driven decisions.
Verdict: True ✅
AI Assistants in Coding
Tools that assist in coding, powered by AI, were predicted to boost productivity by around 30%. This prediction is partially true, as adoption has grown but hasn’t reached universal levels yet.
Verdict: Partaily True 🔄
AI Automation vs. RPA
It was expected that AI automation would completely replace traditional Robotic Process Automation (RPA) platforms. This prediction turned out to be false, as RPA is still widely used alongside AI solutions.
Verdict: False ❌
The final set of predictions dives into specific application areas.
Misinformation and Fake Detector Tools
With the 2024 US elections approaching, it was predicted that major social media platforms would introduce advanced AI tools to detect fake or misleading content. This prediction is true as platforms are indeed enhancing their detection capabilities to combat misinformation.
Verdict: True ✅
AI Media Creation
AI-powered tools for creating text-to-video, image enhancement, and converting images into videos were expected to become widely used, revolutionizing content creation. This prediction is true, as these technologies are now mainstream and transforming creative industries.
Verdict: True ✅
AI in Entertainment
It was predicted that at least one AI-generated concept would result in a full movie, starting from an idea to a screenplay to production. This prediction turned out to be false as there hasn’t been a fully AI-created movie yet.
Verdict: False ❌
AI in Cybersecurity
Both offensive and defensive cybersecurity strategies were expected to rely more on advanced AI, with active participation from governments. This prediction is true as AI is increasingly used in cybersecurity to identify threats and prevent attacks.
Verdict: True ✅
AI App Store Business Model
A new marketplace for AI tools and models, similar to the Google Play or Apple App Store, was expected to emerge, encouraging developers to use foundation models. This prediction is false as no such ecosystem has been fully established yet.
Verdict: False ❌
End Note
The summary of the above predictions is:
✅ 14/20 Predictions: 𝗧𝗿𝘂𝗲
🔄 3/20 Predictions: 𝗣𝗮𝗿𝘁𝗶𝗮𝗹𝗹𝘆 𝗧𝗿𝘂𝗲
❌ 3/20 Predictions: 𝗙𝗮𝗹𝘀𝗲
Srikanth Velamakanni’s 2024 AI predictions highlighted the transformative potential of artificial intelligence across industries. As we enter 2025, many of these forecasts have materialized, with advancements in generative AI, enterprise tools, and AI-driven media creation.
Stay tuned to Analytics Vidhya News for more such updates.
As an Instructional Designer at Analytics Vidhya, Diksha has experience creating dynamic educational content on the latest technologies and trends in data science. With a knack for crafting engaging, cutting-edge content, Diksha empowers learners to navigate and excel in the evolving tech landscape, ensuring educational excellence in this rapidly advancing field.
