2023 was a year of exponential growth in Artificial Intelligence (AI), and it’s safe to say that 2023 marks a turning point in AI innovation. We’ve seen tools that enabled creativity, those that could create text, code and art, and chatbots that engage you in conversation on even the most difficult topics. This year, we’ve seen how AI is transforming the way we interact with technology.
As the year 2023 draws to a close, we’re taking a look at some of the best AI tools that came out in the last 12 months. These tools provide a glimpse into how far AI can go in improving human capabilities.
From image generators like DALL–E 3, which caught people’s imaginations, to advanced language models like ChatGPT from OpenAI and Claude from Anthropic that can communicate with users effortlessly, here’s a look at the variety of innovative AI applications.
ChatGPT
ChatGPT is OpenAI’s flagship chatbot. In November 2022, OpenAI launched ChatGPT, the world’s first chatbot that can answer almost any questions under the sky. It can also ideate, write programs, and more. ChatGPT uses OpenAI’s proprietary Large Language Model (LLM) called GPT-Series. After launching ChatGPT, OpenAI, a start-up AI company based in San Francisco, California, added several new features to its chatbot. It also launched mobile versions of ChatGPT for iOS, Android, and GPT4, OpenAI’s most powerful AI model.
GPT-4 added new capabilities, but OpenAI also gave the chatbot Internet power and added plugins to allow subscribers to use chatGPT for various use cases. At the end of 2018, OpenAI’s first ever developer conference was held and was attended by millions of developers, and ChatGPT’s interface was improved with the addition of customGPTs that let users create their own chatbots. In November, OpenAI was also affected by a shake-up in the boardroom. The company announced that it is planning to launch a GPT store in the next few months.
DALL-E3 has been added to the ChatGPT interface, and the recently launched GPT4 Turbo includes a context window with 1.28K tokens.
Bing AI Chat/Microsoft Copilot
Bing AI Chat is a powerful feature of the search engine Bing. OpenAI may require you to subscribe to their ChatGPT plus to experience its powerful GPT4. Microsoft introduced Bing AI Chat earlier this year. Bing AI Chat uses AI to engage in conversational interactions with users. Supported by OpenAI’s advanced language models, it can perform a variety of tasks.
For example, it can create content, generate ideas, and generate images. Users have praised Bing AI Chat for its human-like conversational capabilities. It can help users write code, plan trips, learn languages, and much more. Now called Microsoft Copilot, the Bing AI Chat is seen as an important step towards integration of AI in search engines, providing an interactive and broad search experience.
Runway Gen-2
Gen-2 is a cutting-edge AI video software from Runway, a Google-backed startup. It has already caught the attention of the entertainment world, as it was used in the Hollywood movie ‘Everything everywhere all at once’ where it was instrumental in creating amazing visual effects.
Gen-2 can generate videos from a variety of inputs, such as simple text prompts, complex images, and even existing videos. In our hands-on review of Gen-2, we found its capabilities to be impressive. The tool has limitations, but the quality of output is impressive.
One area for improvement would be to improve the accuracy of video generation. This is especially true when working with more detailed descriptions or particular image prompts. Improving this level of refinement would make Gen-2 even more powerful when it comes to creating and editing videos.
Dall-E 3
DALL-E (Dictionary Language Algorithm) is an image-generating AI model developed by OpenAI. It was first released in 2021 and OpenAI released its third version earlier this year. DALL-E generates images based on prompts, which are natural language inputs. Currently DALL-E is available through ChatGPT and Microsoft Paint, as well as Bing Image Creator and other services using the API. While these platforms provide similar experiences, the official channel is ChatGPT, which is the preferred way to use the AI model. An OpenAI website post explains that DALL-E ‘is natively integrated into ChatGPT, so you can use ChatGPT as your brainstorming partner as well as a prompt refiner. Just tell ChatGPT your idea, from a short sentence to a long paragraph.’
Midjourney
In short, Midjourney works like a paintbrush guided by words. It is a wildly popular AI tool that creates stunning images based on text description and prompts.
To use Midjourney you need to provide a detailed prompt that describes the scene, style, atmosphere, and mood you want to create. You need to be specific about the colour, texture, and composition, as well as the lighting settings.
Midjourney analyzes these prompts and uses powerful algorithms that combine millions of images and art styles to match your request. Within seconds the AI tool creates unique artwork that you can refine, tweak, or use to create more variations.
In fact, earlier this year Pope Francis in Balenciaga Puffer Jacket went viral all over the internet. Many people believed the images were real because they were so accurate and photorealistic.
