Connect with us


what is Lensa, how to use it and its dangers



Over the past few months we have seen how artificial intelligence has taken over photography. Some of the most outstanding examples have been the DALL-E 2, Stable Diffusion, or Midjourney tools, offering the possibility of generating any type of image from a text description.

For some time now another very popular tool on social networks has come into play: Lensa. The application is available for mobile devices, and the use of AI is more aimed at editing portraits, being able to create amazing works of art from its ‘Magic Avatars’ function. In this article we wanted to explain how it works, all its options, and if it is safe to use.

What is Lensa?

Lensa is a photo editor available for iOS and Android devices. The tool provides multiple options to not only edit the photo itself, but also the appearance of the person in the portraitHence the great popularity it has recently acquired. And it is that in addition to options to change parameters such as white balance, contrast, saturation, exposure and others, users can also modify their facial features, adjusting the size of the eyes, nose and cheekbones, or offering retouching options neck, face shape, and more.

Image: (Instagram)

However, the icing on the cake is its Magic Avatars function, generating portraits close to works of art through artificial intelligence. The photography styles that we can generate from this feature are extensive and very varied, a fact that has led many users to place their photos edited with this tool as their profile image. Hence, the reference to the “multiverse” has become popular with this app, since each image would be like a kind of alternative version of another universe.

How does your technology work?

Really, one of the goals of Prisma AI with Lensa is train your artificial intelligence model through thousands of generations by users. In this way, they will obtain an increasingly precise technology.

The Lensa app makes use of Stable Diffusion technology for the generation of images. However, part of the secret of its success is how comfortable it is to use the app to generate the photos, since the app will limit itself to offering variations of the uploaded images and in multiple artistic styles, so we don’t have to enter any description , something that does happen with tools like Stable Diffusion or DALL-E.

How to create your avatar with Lensa

In order to use Lensa, the first thing you have to do is download its application for iOS or Android. When you start the application for the first time you will have to accept the terms and conditions. Also, you should keep in mind that the application is free, but if you want to make use of the Magic Avatars function, you will have to pay for each set of images you generate, regardless of whether or not you have subscribed to any of the plans they offer. Of course, it should be noted that, if you access the 7-day free trial, you will then have a 51% discount on each of the image packs.

To create your first photos by AI, once you are in the main interface of the application, all you have to do is select the Magic Avatars option and continue with the ‘Try now’ option. We follow the steps indicated by the application, and later will ask us to select between 10 and 20 photos from our gallery.

Lensa App

The same app shows us how the photos should be, considering as good the portraits in which our faces can be seen without any problem. And it is that if we have accessories that cover our face, or other people or objects appear in it, the application will automatically unmark the photos that do not comply with the regulations, so we will have to select other images until we have prepared a total of 10 As minimum.

Once you have selected the photos from your gallery, you will have to specify your gender and then choose one of the image packs. Currently, the app offers the possibility of generating 50, 100 or 200 avatars for a certain price for each set of images. When you have selected the pack, the application will take its time to generate the images in different artistic styles.


Is Lensa free? How much?

As we have mentioned, we can download and use Lensa for free. However, the app has an Unlimited subscription, which offers us to edit an unlimited number of photos, also unlocking all the artistic styles. This is priced at 29.99 euros per yearalthough we can also pay for it weekly or monthly at a price of 2.09 euros and 5.99 euros respectively.


However, even if we unlock its premium features, if we want to make use of the Magic Avatars option, we will still have to pay for each set of photos we generate from your AI. Of course, if we are subscribed, we will obtain a 51% discount on each of the photo packs. We can also access this discount if we access the 7-day free trial of their Unlimited plan, although we must cancel before the trial period if we do not want to be charged. Below these lines we leave you the price of each of its packs without the discount applied:

  • 4.69 euros for 50 photographswith 5 variations and up to 10 styles.
  • 6.99 euros for 100 photoswith 10 variations and up to 10 styles.
  • 9.49 euros for 200 photoswith 20 variations and up to 10 styles.

Is it safe to use Lensa?

After its launch, the application has generated a continuous debate on social networks about aspects related to user privacy and security.

Lensa has been developed by Prisma AI, based in Sunnyvale, California. The company was founded in 2016, and its “star app” Prisma was awarded “App of the Year” by Apple and Google.

ChatGPT, an in-depth guide: what it is, how to use it and the best things you can do with this AI conversational chat

According to the company’s statements for TechCrunch, Lensa makes use of AWS services to process the photos selected by the user. Once the artificial intelligence has been trained on the photographs, the images are immediately deleted, according to the company. However, its privacy policy and conditions leave room for maneuver for the company, so even if the firm has been transparent about the removal of the photos, there may be people who still feel uncomfortable using the app.

If you wish, you can have them remove all your personal information from their servers by sending an email to [email protected] In the event that you have uploaded a photo to social networks, by accepting the terms you have given them permission to use that image for advertising purposes. However, you can rescind this permission by sending an email to [email protected]


It should also be mentioned that, the app collects information about the device, and usage details. In iOS we can reject the collection of data through the device settings. However, Android users do not have this option, so they can contact the company via email [email protected] According to Prisma AI, they collect information for traffic analysis purposes from third-party tools, as well as to improve their service. This data ranges from how we interact with the app, to the number of clicks we make on a link.

So…Is it safe to use Lensa? Well, it is clear that the application collects information of all kinds, just like the vast majority of apps that we use every day, so that depends on how you value the treatment of your data.

However, perhaps the most interesting issue is the fact that we are using the app to train an AI model. This serves to increase the precision of their artificial intelligence, so that, in addition to the surprising results and its benefits, it could also lead to negative situations, especially for the artists and designers union. And it is that more and more companies are making use of AI to generate content, making it increasingly dispense with the work of artists. That is why an enormous debate has been generated on the net around this type of tool.

Continue Reading
Click to comment

Leave a Reply

Your email address will not be published. Required fields are marked *


How OpenAI, the company behind ChatGPT and Dall-E of which Elon Musk is a founding partner, makes money



If there is a technology company that has given rise to talk in recent months (with the permission of Elon Musk’s Twitter), that has been OpenAI: the possibility that anyone could try the revolutionary conversational chat for free (some compare that initial approach to the one that he had when he met the internet) of ChatGPT or the creativity of DALL -E2 are only the tip of the iceberg of his potential. But it is that OpenAI was presented to the world as a non-profit artificial intelligence research company. Where does OpenAI get the financial muscle for the research and development of its tools?

What is Open AI?

First of all, a little introduction: OpenAI is a research and development technology company in the field of artificial intelligence. In his own words, his mission is “ensure that artificial general intelligence benefits all of humanity” .

Among his releases, in addition to numerous papersthey find each other three products more or less known throughout the world: the Trained Language Generator (GPT) model, on which ChatGPT and its different versions are based, the product learning model that generates images from text prompts Dall-E and Whisper, an open source neural network that automatically convert audio files to written text.

Models developed by OpenAI are available through APIs or as open source projects, so the code is freely available.

Who is behind OpenAI? Sam Altman, Greg Brockman, Ilya Sutskever, Wojciech Zaremba and the illustrious Elon Musk began this journey in 2015.

How OpenAI makes money

In the intro we anticipated that OpenAI began as a non-profit organization. But there is an important nuance: as of 2019 it became a “limited partnership” (LP) – a type of company in the United States – something like a limited company. However, according to OpenAI this LP continues to be supervised by the non-profit organization. This change in its constitution is of vital importance, both for access capital to earn money.

There are three ways that OpenAI makes money: through fees to access its models, through subscriptions, and through investments.

  • rates For those who use their models professionally, here are the rates. While image models have a unit price per image based on resolution, language models are priced in dollars per piece of text (tokens), although prices change depending on the language and its capabilities and whether you choose to customize existing models.
  • subscriptions. In a few weeks of free and public access, the use of ChatGPT has become so popular that it is already used to answer questions, for classes or even to flirt on Tinder. But in just a few weeks some paid versions have already appeared, such as the fleeting ChatGPT Professional oriented towards a more prowith a cost of 42 dollars a month and advantages such as availability even in periods of high demand, faster responses and priority access to new features that are being launched or ChatGPT Plus “for the impatient”, more affordable than the previous one and also faster and more prioritized over the free version.
Screenshot 2023 01 30 At 9 58 45

Interface to hire the payment option. Via Reddit.

  • Investments. Within OpenAI there is a department launched in 2021 called “OpenAI Startup Fund”, whose objective is to invest in startups in the artificial intelligence sector. The budgeted size of the investment is $100 million and, in the words of Sam Altman, “it’s not a typical venture fund. We plan to make big early-stage bets on small companies, probably no more than 10.” There is no data on the return of these investments, but it is normal to receive shares in the form of shares that can later be resold at a higher price.

What about Microsoft? Presumably other fees may apply, given its ongoing relationship with OpenAI since 2019, at which time it invested over $1 billion. The economic injections have been repeated in 2021 and a few days ago and are estimated at billions of dollars. During this time, Microsoft has provided processing units necessary for model training, has designed and built a supercomputer, and puts its computing services at the service of OpenAI. In return, it starts from a privileged position to integrate OpenAI’s advances in artificial intelligence into its products, as is the case with the arrival of GPT to Azure and Office365.

Let’s talk about money. According to The Information, OpenAI is currently valued at twenty billion dollars. According to Crunchbase, OpenAI has managed to raise more than eleven billion dollars in six investment rounds, the last one dating from January 23rd. If we talk about investors, the aforementioned Microsoft is one of the main players, but there are also others such as Sequoia Capital, Tiger Global Management or Andreessen Horowitz. Although there are no revenue figures for the company, Reuters reports that in different presentations to investors it has stated that expects to enter 200 million in 2023 and this figure will rise to one billion in 2024.

Being “free” is also a business with benefits

Beyond these three clear ways in which OpenAI makes money, it is worth stopping and delving into its business model. Regardless of your business products, the goal is to become a a platform from which other businesses and services are built and developed.

ChatGPT and DALL-E are two models available through the APIs that generate income from fees derived from their use, it is clear. But the fact that they exist for free brings non-economic benefits to OpenAI, such as training the algorithms themselves at zero cost and passively or simply, making themselves known to the general public. Take this tweet from Sam Altman as an example: it took just five days for ChatGPT 3.5 to reach one million users.

Its popularity was reinforced with experiments on social networks, users looking for tickles and controversies such as how education will change with such a powerful tool for cheating within everyone’s reach. They say that all publicity is good and here it applies: It is not the first conversational chat that exists, but it is the most mainstream. In fact, It is the platform that has grown the fastest in the entire history of the internet, above social networks such as Instagram or TikTok. In the case of DALL-E2, the images themselves invite virality due to their style or jokes about the excess of teeth or fingers generated in images from text.

In other words, using OpenAI’s artificial intelligence models for free makes your tools more powerful and more visible, two very useful features to call attention to potential business clients.

An important point of OpenAI is that it does not allow its clients to export their debugged models, generating a kind of symbiosis: the more the companies that use its tools grow and their requirements increase, the more they will spend (and consume resources) on OpenAI; on the other, OpenAI needs technical and financial muscle to increase its computational power (GPT-4 will have 100 trillion parameters, almost 600 times more than its predecessor). It also needs more people to do it, so at a time when most big techs are laying off, OpenAI hires. This growth also allows us to glimpse a potential scenario that we cannot rule out in the medium term in a growing segment: that it may acquire other products from the competition to continue strengthening.

Photography | Own montage and Photo by Alexander Gray on Unsplash

Continue Reading


How to fix WiFi problems in Windows 11 and its most common failures



At present it is really strange to have a computer that does not connect to the internet to be able to work on the network, and sometimes whether it is a desktop or laptop computer we do it through WiFi. It is known by all that the most recommendable is to make this connection through the cable, but sometimes due to the location of the router or simply the fact of having to be on the move, the wireless connection is used.

But just like with other features, WiFi can fail on more than one occasion and make it impossible to connect to the internet without a connection. This is one of the most common problems when we talk about a computer that owns Windows 11. In this article we analyze the most frequent errors and how to solve it.


Check if the fault is in the router itself

One of the most common problems in relation to the Internet connection is in the router itself. That is why the first thing to do is connect to a network of a different frequency. You have to know that many routers have two different network bands: 2.4 GHz and 5 GHz, and this is how it will appear in the list of networks. To do this, you will have to switch between them to check if you can have an internet connection or not. This is mainly because the 5 GHz network is recommended for when we are close to the routerbut if the computer is far away it is better to opt for the 2.4 GHz one even if it is more saturated.

Obviously another of the classics within the revision of the router is the fact of restarting it to be able to create a new connection with the internet service provider. To do it correctly, the following steps must be followed:

  1. Disconnect the power cord from the router.
  2. Wait 30 seconds minimum.
  3. Reconnect the router to the power source.
  4. Please wait a few minutes until all the LED lights turn on and the connection with the devices is made again.

At this time, you will be able to try to connect to the internet again with your computer. If it still doesn’t work, it’s important Check if the connection can be made with the rest of the devices in the home, such as the mobile. If so, then the fault will be in the computer hardware or in the software itself with Windows 11.

How exactly does a speed test work and how reliable are they?

Detect where the problem is

With Windows there are ways to know where the problem of the internet connection may be. Specifically through the command prompt you can perform a ping test that will allow us to know if the computer gets a response from the router and this may indicate that the problem is with the provider itself. To do this, the following steps must be followed:

  1. Open the Start menu.
  2. In the search engine type “command prompt” and run it as administrator.
  3. Type ipconfig and write down the IP address that appears next to Default Gateway. Normally it will be
  4. Again in the CMD type ping and press Enter. The command example would be ping
  5. If you get a response with the phrase “Response from bytes=32 time=5ms TTL=64″ it can be said that the problem is with your system provider. But if you do not get any response, you will have to take different steps to solve it.

Run network commands to reset the network

In computing we always say that the most logical step to solve any problem is to restart the system. And precisely the same can be done with the network connection to the be able to reset all its parameters by renewing the IP address and clearing and resetting the cache of the DNS client resolver. Personally, this is one of the steps that has always worked for me when I have had problems in Windows 11.

In order to carry it out, they will have to enter different commands in CMD and then check if the internet connection has been restored. The steps to follow will be the following:

  1. Open the Start menu.
  2. In the search engine type “command prompt” and run it as administrator.
  3. In the window that opens, type the following commands in the order marked:
  • netsh winsock reset and press Enter.
  • netsh int ip reset and press Enter.
  • ipconfig /release y press Enter.
  • ipconfig /renew and press Enter.
  • ipconfig /flushdns and press Enter.

Uninstall adapter driver and reboot

In the event that these steps did not work correctly, it is important to uninstall the network adapter driver. This is responsible for offering the precise instructions to be able to correctly receive the Internet connection by the router. But sometimes it may end up failing and the smart thing to do is to uninstall it and force Windows to reinstall it. It is important to previously locate the driver on the internet with another computer and have a backup copy on a flash drive. This is because if the failing PC is offline, you won’t be able to reinstall it.

How to fix problems or failures when installing updates in Windows 11

In order to carry out this process, you will have to follow the following steps:

  1. Open the Start menu.
  2. Writes device administrator and select Device administrator in the listing.
  3. In the list look for network adapters and find your adapter.
  4. Right-click on it and click Uninstall device.
  5. Reboot the PC.

When the device starts up, and if all goes well, this driver will need to be reinstalled and the internet connection will be restored. But if you haven’t installed it, it’s important rescue the backup that has been made previously.

The bug may be in the latest Windows 11 update

It is a fact that sometimes the updates that come from Microsoft sometimes give more bugs than solutions. You can know if this is responsible if the failure has been filed right after performing a Windows 11 upgrade. In this way, you should choose to temporarily uninstall the latest update that has been installed through Windows Update. The steps to follow are those:

  1. Tap Start.
  2. Scroll down to Settings.
  3. On the side, click on Windows Update.
  4. Click on Update history.
  5. Select in Uninstall updates.
  6. Select the latest installed update and tap on uninstall.

Check for driver updates

In the event that uninstalling the latest Windows 11 update has reestablished the internet connection, you will have to search for updates to the driver itself. This is something that is done by following these steps:

  1. Open the Start menu.
  2. Writes device administrator and select Device administrator in the listing.
  3. In the list look for network adapters and find your adapter.
  4. Right-click and click Update driver > Search for updated driver software automatically.
  5. If the update is found and installed, restart your PC.

If no type of new driver is found, it is important to also visit the manufacturer’s website and track down the version numbers they have released and check them with the ones you have installed on your computer.

reset network

The last step to take into account is the restoration of the network, which is a radical decision when remove any network adapters that have installed your configuration. Rebooting the computer will reinstall the network adapter. That is why it is an ideal solution when upgrading from Windows 10 to Windows 11 as this is a common failure in this situation. The steps to follow are those:

  1. Tap Start.
  2. Scroll down to Settings.
  3. On the side, click on Network and Internet.
  4. Click on Advanced network settings.
  5. Choose network reset and confirm the action.
  6. It is important to restart the computer immediately so that the adapter is reinstalled and check if that has solved it.
Continue Reading


Working four days a week reduces the stress of workers and this gives more income to companies: so says this study



In the year 2022, the interest in working four days a week skyrocketed like never before, also in European countries and in Spain, where the government is preparing aid for companies that want to test this work model, after an idea promoted in Congress by Más País.

We now have the results of a large study conducted by researchers from Cambridge Universities, Boston and Dublin. 33 companies were analyzed. Some of the companies gave their workers every Friday off, while others did not have a common day off.

The investigation started a year ago: in February 2022.

In general terms, the study has positive conclusions: the reduction of the working day has the potential to bring social, economic and climate benefits, social, economic and climate benefits (less commutes to the workplace). When is accompanied by an increase in hiring, it can reduce unemployment. It can also reduce costs in situations where employees experience high levels of stress.

The companies have been very happy

On the part of the company, the experience “has been a success”, say the conclusions of the report. Of the 33 companies 27 filled out a final survey in which they were asked about their experience and if they would continue with the four-day week.

On a scale of 0 to 10, from very negative to very positive, the average company rating of companies is 9. Asked specifically about productivity, companies gave a score of 7.7. Of the 27, 18 are definitely determined to maintain this formand no one is opposed to continue.

In addition, on average the income of the companies increased more than one
percentage point per month during the trial. The increase in revenue was accompanied by a decrease in the number of employees in the participating companies. On average among the 18 companies that provided data on this parameter, there was an increase of 12.16% in the number of employees.

Workers’ conclusions

From the point of view of the workers, on a scale of 0 to 10, from very bad to very very good, the mean overall test experience was 9.1. 97% of employees want to continue the trial.

I was a digital nomad when nobody else was: teleworking was a lot of fun, stressful (because of technology) and nobody understood what I was doing


There is a wide range of well-being indicators: stress, exhaustion, fatigue and family conflicts decreased, while physical and mental health improved. Employees used their day off for hobbies, housework, and personal care.

Still, people put in overtime. Specifically, from 40.83 weekly hours they went to 34.83 and not 32. However, there was a reduction of 6 hours of work. Job stress decreased during the trial period. On a scale of 1 to 5, from never stressed to always, before starting the test people said, on average, that they were stressed at 3.15. This was reduced to 2.95.

Little reduction, although twice as many people reported feeling less stressed. Exhaustion also decreased.

Continue Reading