I tested six AI chatbots by creating a WordPress plugin. Find out if ChatGPT, Bard, Bing, Claude, Code Llama, and Llama 2 completed the task.
Generative AI could fill knowledge gaps, allowing people to experiment with new projects.
To test the coding abilities of the top AI chatbots, I tried creating a simple WordPress plugin.
Four out of six AI chatbots completed the task without requiring revisions.
Artificial intelligence (AI) chatbots have opened up a world of possibilities for people to dabble in projects they might not have tackled due to lack of experience.
As someone who uses WordPress but doesn’t have coding experience, I thought it would be interesting to see which AI could help create a simple plugin.
For this little test, I used the following six AI chatbots:
GPT-4 for ChatGPT Plus subscribers.
GPT-4 via Microsoft Bing AI.
PaLM 2 via Google Bard.
Claude 2 via Anthropic’s website with a free account.
Meta’s Code Llama (codellama-34b-instruct) and Llama 2 (llama-2-13b-chat) via Perplexity Labs with a free account.
While not unique to WordPress’ repository of plugins, I submitted the following prompt to describe the desired functionality.
No Revisions Required For ChatGPT, Bard, Bing, And Claude
ChatGPT, Bard, Bing, and Claude generated PHP code that created no issues during the new word count plugin’s installation, activation, or deactivation.
Of the four, ChatGPT provided the cleanest set of directions on installing the newly generated plugin, followed by Bing and Bard.
Llama Requires Assistance From Other AI Chatbots
While codellama-34b-instruct in Perplexity Labs generated the PHP code for a WordPress plugin, it offered a few additional challenges.
After installation, the plugin offered a series of warning errors instead of word counts.
When I asked it to fix the problem, it offered suggestions.
Generative AI And The Future Of Coding
Generative AI holds immense promise for various development tasks, including creating a simple WordPress plugin.
However, not all AI models are created equal. Some stood out for their ability to create effective code with minimal prompting and troubleshoot code generated by other models.
With the rapid development of AI technology, it shouldn’t be long before other AI models become more reliable and accurate in accomplishing coding tasks.