Generating videos with artificial intelligence from text can be done through several tools. One of them is from Luma Dream Machine, a page that can be used for free and only requires one email account to start creating audiovisual pieces.
To use Luma Dream Machine, follow these steps:
1. Enter the Luma Dream Machine website.
2. Sign in with a personal email account.
3. Enter the text or prompt of what you want the video to be like in the white box in the center of the page.
4. Select the arrow and wait for the video to be generated.
5. The video generation process takes approximately two minutes. To view the video, just go to the My Videos section.
Luma does not specify if there is a limit of videos but it does mention that an account may be suspended if the user requests the generation of a piece with adult content. or deepfakes, for example.
Luma collects the following information from its users:
– Contact information such as first and last name, and email address.
– The communications exchanged with users, including when communication is established through our website, applications or on Discord.
– Payment information required by our payment processor, Stripe, Inc., “We do not store payment card numbers,” they say.
Luma also collects images and information related to users’ advertising preferences. The company points out that this data is shared with:
Service Providers: Luma may share this personal information with third party companies and individuals who provide services on their behalf or help them operate their services.
Professional advisors: Luma has the ability to disclose this data to professional advisors, such as lawyers, bankers, auditors and insurers
“We may share your personal information for compliance, fraud prevention and security purposes described above,” they add.
Some users on X (formerly Twitter) pointed out that Luma Dream Machine is a replacement for Sora, OpenAI’s AI that also generates videos.
Sora is not available to the public but it is one of the most popular tools of this style because it is from the same company that is behind ChatGPT.
The company, led by Sam Altman, says Sora can make videos that look very real and produces exactly what users ask for, without losing quality in the images.
Currently, Sora is being used only by an exclusive group of people belonging to the OpenAI “red team”, with the main objective of detecting possible errors, damages or risks in the tool.
The “red team” is a specialized group whose main task is to evaluate and test systems, software or applications to discover vulnerabilities, flaws or potential risks. Their job is to think and act like potential adversaries, simulating the methods and tactics that a real attacker might use.
Likewise, OpenAI has given access to Sora to a small community of visual artists, designers and filmmakers in order to collect their opinions and improve the model, adapting it to be more useful in the creative field.
The company has chosen to share the developments and advances of its research from the early stages to establish a feedback channel with people outside of OpenAI, allowing for a better understanding of how the capabilities of artificial intelligence can be enhanced for benefit and progress in various areas.
However, the company has warned that, like any other AI tool, Sora has its own limitations and challenges.
For example, Sora might have trouble accurately representing physics in complex scenes or fail to identify specific relationships prompts in a video.
