guess who created the logo
Try to distinguish logos created by artificial intelligence from the work of human designers.
Today, there are various services based on artificial intelligence in the public domain, for example, Looka, Midjourney and DALL·E 2. They are able to design a logo or illustrations, as well as create entire art objects.
I decided to test in practice whether it is possible to distinguish neural network logos from the works of well-known branding agencies: check your intuition, and then click on the “spoiler” to find out the right answer.
Logo for a technology startup development studio
hidden text
This is the work of the Logomachine 🙂
Logo for a family farm
hidden text
this is the work of the neural network midjourney
Most neural networks so far only work with a raster. The logo from Midjourney is no exception, it needs vector refinement in order for the brand to use it without loss of quality.
Logo for the importer of wines and spirits
hidden text
Logo by Depot branding agency
Logo for a recruiting company
hidden text
This is another work of the Logomachine 🙂
Logo for a programming studio
hidden text
This is the work of the neural network of the Art. Lebedev studio – Nikolai Ironov.
In my opinion, in this case, the neural network did not finalize the adaptive variants of the logo. In the future, the complex shape and large number of small details may create problems when scaling it.
Logo for a brand of women’s clothing
hidden text
This is the work of the neural network looka.
Since the neural network does not take into account the wishes of customers, the final result is always unpredictable. It can generate both bold and, as in this example, sometimes banal design.
Logo for time tracking service
hidden text
This is the logo of the Art. Lebedev studio (Nikolai did not participate in its development).
Logo for video calls app
hidden text
This is the work of the neural network looka.
The neural network did not take into account the combination of the logo with the media. If a brand uses several patterns, it will need to contact designers for additional corporate identity development.
Hat factory logo
hidden text
This logo was developed by Suprematika branding agency.
Secret logo
hidden text
Reddit user decided to test the neural network DALL E 2: the logo is free, if you are interested, get in line!
PS Neural networks do not always cope with image detail. In this version, when you zoom in on the bull, you will see its broken and asymmetrical contours.
Can neural networks be considered a full-fledged replacement for designers?
While neural networks are at the development stage, full-time specialists can sleep peacefully. Now designers use artificial intelligence as an applied tool, after which they need to finish the work manually, or limit it to templates. Neural networks have great potential, but with their pros and cons.
pros
Neural networks do not get tired of working: if designers create two to five logo options for a client to choose from, then AI is able to generate several thousand images in 5 minutes.
Neural networks do not need to spend a lot of time searching for creatives, relying either on their own observation or on graphic resources in the public domain. AI is the sum of the experience of thousands of designers with an endless source of ideas to implement.
Minuses
Neural networks cannot evaluate their own design: they do not have indicators to help distinguish bad design from good. The decision is shifted to the customer – the neural network generates options, and the person chooses based only on his expertise.
Neural networks still do not work well with external media. They turn the logo into a pattern, ignoring the texture, volume and characteristics of specific objects. At the moment, only humans can work on this task.
In short, if a futuristic and modern logo is within the power of artificial intelligence, it has not been trained in deep elaboration of corporate identity.
Tell us in the comments how many correct answers you could guess!
