I adore my backyard and have a passion for gardening. These sentiments do not appear to be inherently contradictory. A well-maintained garden is not only visually appealing but also gratifying. However, my lack of expertise in creating a sustainable and manageable landscape makes this task quite labor-intensive and demanding in terms of ongoing upkeep. Interestingly, ChatGPT seems to excel in gardening and displays a keen interest and proficiency in this area.
Artificial intelligence (AI) can be likened to the existential query posed by Billie Eilish, “What am I here for?” in the realm of current technology. While the possibilities with AI are vast, the outcomes are often contingent on the input provided. I have dedicated significant time to exploring AI applications in writing, computing, and interactive conversational interfaces. While initial results are promising, sustaining the performance over time proves challenging. Some of my initial experiments date back a month or more, which in the realm of AI equates to years.
Recently, I have delved into experimenting with the latest large language models (LLMs) and image generation tools such as Microsoft Copilot, Google Gemini, and OpenAI’s ChatGPT. Transitioning predominantly to mobile platforms, I discovered that a mobile AI gardening assistant serves as the landscape consultant that was conspicuously absent when I first engaged with early AI chatbots on desktop interfaces.
Utilizing ChatGPT Plus, the subscription-based AI service offering access to GPT-4 and DALL-E 3, for $20 a month, proved to be a game-changer. GPT-4 distinguishes itself by being trained on more recent data compared to its predecessor, GPT-3.5. The incorporation of up-to-date information could potentially enhance gardening advice significantly. For instance, insights on current weather patterns might facilitate recommendations tailored to my specific climate, rather than relying on outdated data. OpenAI’s approach of training its large language models by aggregating vast amounts of online data, including publicly available weather information, underscores the potential for more accurate and relevant guidance.
While my rear and upper gardens are in decent condition, the front yard poses challenges. This area, receiving minimal sunlight, has proven inhospitable to many plants. Seeking ChatGPT’s expertise, I provided details about my location (northeastern US) and the prevailing temperate climate with moderate rainfall, prompting suggestions for landscaping solutions. Leveraging ChatGPT’s versatility in processing text, voice, and visual inputs, I began by capturing an image of the problematic area for analysis.
ChatGPT adeptly identified the plants in a lucid and conversational manner:
- A striking red rose amidst verdant foliage.
- Daylilies encircling the rose with their clean, strap-like leaves.
- Rhododendrons, characterized by their colorful flowering shrubs.
- A cluster of white flowers with yellow centers.
- A low-growing evergreen leaf resembling either juniper or creeping rosemary.
Subsequently, ChatGPT proposed a selection of flowers suitable for the setting, including coneflowers, black-eyed Susans, shrubs, bee-balm, emphasizing their perennial nature. Consulting with my family, who possesses a deeper understanding of plants and foliage, confirmed the viability of these choices.
For the latest updates, exclusive offers, and more from the TechRadar group, subscribe to our newsletter.
Beyond mere question-and-answer interactions, the true value of utilizing an AI bot lies in the engaging dialogue it fosters. When I inquired about the suitability of these plants considering the limited sunlight exposure in that area, ChatGPT reassured me by affirming that the recommended species typically thrive in conditions with 4-6 hours of direct sunlight daily. Furthermore, it elaborated on how each plant adapts to varying light levels, instilling a sense of reassurance.
Impressed by ChatGPT’s proficiency, I decided to test its plant identification capabilities further. Capturing an image of a hawthorn growing near my window, I queried ChatGPT about the plant type and the optimal care practices in my locale. Promptly, ChatGPT identified the plant as a budding tree, indicating that the photo was taken in early spring. Its accurate assessment and guidance provided valuable insights without disparaging my gardening skills.
Embarking on more challenges, I presented ChatGPT with a realistic-looking artificial bouquet, which it promptly identified as artificial due to the lack of variation in color and form typical of real flowers. This attention to detail and discernment showcased ChatGPT’s keen observational skills.
As I contemplated mulching my rose bushes and tree beds, I sought ChatGPT’s advice on the timing and type of mulch required. While the response was somewhat vague, ChatGPT recommended a mulch layer thickness of 2-3 feet, along with a comprehensive overview of mulch types and their respective benefits and drawbacks. Although unable to physically assess the mulch depth, ChatGPT’s guidance aligned with conventional wisdom shared by experienced gardeners.
Addressing the challenges in my backyard, particularly the patchy and thinning grass, I sought ChatGPT’s expertise on rejuvenating the lawn. ChatGPT astutely identified the underlying issues such as soil compaction, nutrient deficiencies, and pest infestations, offering practical advice on aeration and pH adjustment using lime. This demonstrated ChatGPT’s ability to provide tailored recommendations based on specific environmental conditions.
Inquiring about the ideal grass seed for my climate, ChatGPT promptly furnished a concise list of five suitable options. While ChatGPT does not explicitly cite its information sources, it generously shared general gardening advice tailored to my location, with citations from reputable sources like Finegardening.com, Savvygardener, and the US Government.
Exploring options for a small flower bed adjacent to my house, ChatGPT recommended strawberries among other plants. This suggestion resonated with me, as strawberries had previously occupied that space before succumbing to unfavorable conditions.
Despite its proficiency in providing textual advice, ChatGPT encountered challenges when tasked with generating landscape images based on its recommendations. The resulting images, while imaginative, deviated significantly from the original photo, featuring fantastical landscapes and structures that bore little resemblance to reality. Nonetheless, the practical advice offered by ChatGPT outweighed the limitations in image generation.
In conclusion, ChatGPT emerges as a reliable and knowledgeable gardening companion, offering tailored advice, plant recommendations, and valuable insights to enhance my gardening endeavors.
