What Are the Latest Trends in Artificial Intelligence and Machine Learning?

Key Points

• Creative AI models like ChatGPT and Midjourney are gaining prominence and refinement, enabling more natural language communication and expressive art generation.
• AI and machine learning are increasingly used to enhance productivity in fields like agriculture and electronics, with autonomous tools and quality inspection models.
• Ethics and regulation in AI and ML are hot topics, with calls for more active regulation and potential controversies surrounding data collection.
• Low code or no code AI solutions like Amazon Sagemaker, Akkio, and Apple CreateML are gaining traction, allowing non-developers to leverage AI and ML models.
• AI voice generation is improving, with more realistic-sounding text-to-speech models, but ethical considerations and refinement are still needed.

Artificial intelligence and machine learning have been in the news quite a bit lately. As businesses and individuals alike move to embrace this rising technology, it is essential to keep an eye on developments.

AI is a relatively young field, at least in regard to usable products, but there is great potential for many uses. That said, as with any field of development, there is movement and growth as things become easier and the resources more plentiful.

Artificial intelligence and machine learning are here to stay, and developers are exploring the options available. So, with that in mind, let’s take a closer look at some of the rising technologies and trends present in AI and ML development.

Creative AI

ChatGPT is one of the more prominent examples of creativity in artificial intelligence and machine learning. There have been recent strides, especially with the recent release of GPT-4, in creating AI language models which can more effectively communicate in a natural and human manner.

GPT-4 is capable of generating musical notation, short stories, and other written works. It lacks the general flow of human language, as you can see in some of its content.

Other creative artificial intelligence models like Midjourney have gained massive prominence, thanks in part to its ability to generate expressive works of art with simple prompts. The works themselves can be goofy, breathtaking, and everything in between.

Midjourney, much like GPT-4, lacks some of the general refinements you might see with a practiced artist. This comes to the fore particularly when looking at how it renders anatomy.

ChatGPT and Midjourney gained prominence in 2022, but 2023 has seen both models gain further refinements and utility. It will be fascinating to see how they develop and grow refined with further work behind them.

AI Human Collaboration

Artificial intelligence and machine learning have both seen growing prominence as a means of increasing productivity for workers. This is seen particularly in a variety of fields like agriculture and electronics.

Here, productivity increases thanks in part to the use of technology. Machine learning models can be trained as quality inspectors for printed circuit boards and other vital electronic components. This cuts down on the team needed for actual human intervention.

In agriculture, autonomous tools are being seen in use for the actual planting and cultivation of crops. Autonomous drones and the like help spare farmhands from some of the more backbreaking aspects of their careers and enable efforts to be placed elsewhere.

As the technology behind these autonomous co-workers grows more robust and refined, you can expect to see them in more and more fields. By eliminating some of the more tedious parts of the job, workers can work in areas that need more human intervention.

Artificial Intelligence and Machine Learning Ethics and Regulation

Ethics and regulation have been hot-button topics for the usage of AI since its emergence in Zeitgeist. Regulations and ethical considerations for artificial intelligence and machine learning are almost a given. When you consider language learning models like ChatGPT and how they source their material, it does raise certain ethical concerns.

Recently, the CEO of OpenAI called for a meeting with the United States Congress to call for more active regulation. When considering the usage of AI for malicious purposes, it is almost a given that some changes are due.

This has led to some controversy, however, as CEO Sam Altman has threatened to leave the EU if not allowed to operate. The EU, in particular, has proposed sweeping reforms, thanks to the somewhat nebulous way in which data is collected for these publicly available AI models.

Low Code or No Code Artificial Intelligence

Low code or no code AI refers to users not needing an exhaustive background as a developer to leverage an AI or ML model. This is starting to gain prominence in fields where a background in software development isn’t needed.

When considering the usage of AI in other fields, like medicine or law, it makes sense to have artificial intelligence present to help with some of the nitty-gritty details of the job. Your average lawyer isn’t a software developer by trade, however.

This is where no code or low code artificial intelligence comes in handy. Products like Amazon Sagemaker, Akkio, and Apple CreateML offer creative solutions without the need for a background in programming.

As such, once implemented in an organization, these products can be bespoke and tailored to the needs of a business. While hardcore programming is still needed for more nuanced approaches, like autonomous driving, it isn’t necessary for a virtual paralegal.

Artificial Intelligence and a Rise in Cybersecurity Needs

Cybersecurity has become a crucial part of the modern world. As such, there are some concerns as to how artificial intelligence and machine learning models can be implemented maliciously. ChatGPT can generate blocks of code, which can lead to bad actors generating malware at an astonishing rate.

The rise in security concerns isn’t unfounded, either, as a recent article from Gizmodo explored how adept ChatGPT is at creating malware. Now, the recent GPT-4 is supposed to restrict malicious requests, but it isn’t the freely available one.

In cybersecurity, there is malware that is deemed polymorphic. What this means is that a virus or worm changes its approach so as to avoid detection. Someone with a little know-how on how to generate the right prompt could wreak havoc on an organization.

As such, it is important for cybersecurity professionals and teams to act proactively. The usual best practices for confidentiality, integrity, and accessibility still apply.

AI Voice Generation

Text-to-speech is nothing new, but having a more human-sounding response certainly is. Firms like LOVO and MURF.AI are leveraging artificial intelligence and machine learning to create more realistic-sounding text-to-speech models.

There is still some work to be done in this field, but the results are already quite astonishing. Users can create facsimiles of popular figures to act as the guiding voice for their AI model.

It has also recently caught quite a bit of controversy, thanks in part to the usage of this technology to perform as popular musicians. Spotify recently just purged a whole host of accounts from their streaming service.

Artists like Drake have experienced the usage of AI voice models to copy their distinctive vocal flow and delivery.

It certainly has more mundane uses, where AI voice models could be more inviting than something akin to Siri or Alexa. More refinement and ethical considerations need to be in place before AI voice generation is ready for prime time, however.

Autonomous Driving and the Automotive Industry

Artificial intelligence has had its place in the automotive industry for a number of years. While its inclusion isn’t a new phenomenon, great strides have been taking place to implement AI in newly manufactured vehicles.

When you consider manufacturers like Tesla and Volvo, it is easy to see where this technology fits into place. Artificial intelligence and machine learning represent not only ease of manufacturing but also provide greater safety measures for drivers on the road.

AI models implemented into a vehicle’s operating system can provide visual detection for objects and pedestrians, and even regulate the speed of a vehicle on the highway. Autonomous driving is also another very popular use for AI models.

Self-driving cars have been present for a while. However, greater refinements in AI models might see viability for them in city streets in the near future. Autonomous models require constant high-powered AI usage to visually discern between objects, navigate roads, and ultimately avoid accidents.

Visual annotation techniques for the artificial intelligence models powering autonomous vehicles have seen great strides just this year. While it is a frightening use of the technology at first glance, there have been growing calls for open standards for safety and speed of development.

What’s Next?

AI and machine learning will continue to see more developments as time goes on. For a comprehensive overview of AI and ML trends to watch out for, we recommend the following video:

The image featured at the top of this post is ©robypangy/Shutterstock.com.