The AI Year in Stories: Spring 2024

December 24, 2024

undefined mins

Anthropic states that the Claude 3 family will be available at a lower cost than competitor models

We look at the articles that made the news in AI in 2024. Today, it is March, April and May's turn, featuring Claude's new model and Tech Layoffs

Anthropic Unveils Claude 3: Its Most Powerful AI Chatbot Yet

AI start-up company Anthropic announces Claude 3, a new family of AI models that offer a wide range of new capabilities.

Models Opus and Sonnet are now available, with the Haiku model expected to be made available soon. Each model aims to offer a more powerful performance and allow users to select the optimal balance of intelligence, speed and cost for specific business use cases.

Significantly, Anthropic states that the Claude 3 family will be available at a lower cost than competitor models currently on the market. Unlike earlier versions, Claude 3 is able to understand text and photo inputs, making it multi-modal.

New features to enable larger-scale AI deployments

Claude 3 comes with a range of features, including answering more questions, understanding longer instructions and being more accurate. Given that the models can understand more context, Anthropic says that they can therefore process more information.

The largest and most intelligent model, Claude 3 Opus, offers high levels of performance when it comes to highly complex tasks. Utilising the power of generative AI (Gen AI), Opus can navigate open-ended prompts with remarkable fluency and human-like understanding.

Also available is Claude 3 Sonnet which delivers strong performance at a lower cost and is engineered for high endurance in large-scale AI deployments. Ultimately, these models are designed to be easier to use and simpler to instruct for use cases like natural language classification and sentiment analysis.

Anthropic also highlights that the new models will be an improvement to its previous model, Claude 2.1, with Sonnet in particular operating twice as fast and able to excel in rapid-response tasks like knowledge retrieval or sales automation.

“In our quest to have a highly harmless model, Claude 2 would sometimes over-refuse,” Anthropic co-founder Daniela Amodei told CNBC. “When somebody would kind of bump up against some of the spicier topics or the trust and safety guardrails, sometimes Claude 2 would trend a little bit conservative in responding to those questions.”

“We’ve tried very diligently to make these models the intersection of as capable and as safe as possible.”

Tech Layoffs Rise yet Businesses Seek to Plug AI Skills Gaps

Despite concerns over AI and automation replacing humans, it is human skills that are still integral to enterprise development

Large technology companies are continuing to lay off swathes of employees, contributing to public anxieties about the power of artificial intelligence (AI).

Microsoft, Amazon, Google and Dell are just some of the big tech organisations that have committed to letting more than 57,000 members of their combined workforces go in 2024 alone, with more expected in the coming months.

This compares to nearly 1,200 tech companies laying off 263,180 people in 2023. Reasons given were rebalancing from the hiring boom in the wake of the COVID-19, increased competition in an explosive digital market and restructuring business areas partly to prioritise AI.

Despite this, companies remain in need of highly skilled workers to train and deploy their AI models.

A need to build strong AI teams

Some have blamed the job cuts on AI, suggesting that businesses are considering emerging technologies to ‘plug gaps’ in their strategies, increasing efficiency and reducing costs. In a recent survey by McKinsey, 25% of business professionals said they expect job cuts within their workplace on account of increased AI adoption.

However, it can be argued that these organisations are instead seeking to build more specialised teams that can handle AI better.

New research from Advania, a Microsoft partner, reveals that the complexity of IT systems is delaying actual progress when it comes to harnessing new technologies to their full potential. It cites that 81% struggle to scale, update and future-proof their underlying tech stacks.