Behind the curtain, Builder AI's development process relied on around 700 human engineers in India. These engineers manually wrote code for client projects, while the company portrayed the work as AI-generated.

Contrary to its claims, Builder AI faked its AI capabilities, employing a large team of engineers instead of using true artificial intelligence. The company raised over $450 million at a valuation of $1.5 billion from investors like Microsoft, Qatar Investment, and SoftBank, based on these deceptive practices.

Furthermore, Builder AI worked with a company called Verse Innovation to routinely inflate revenue figures through a practice known as "round-tripping." This allowed them to present inflated financial data to investors.

Ultimately, Builder AI's creditors caught on to these fraudulent business practices and seized the company's accounts. As a result, Builder AI is now filing for bankruptcy. This scandal highlights the risks of hype and overvaluation in the AI industry, where some companies may resort to deception to attract investments.

Luma AI's modify video feature showcases impressive capabilities, allowing users to reimagine videos in different styles and even insert characters while maintaining consistency. The demos on their website are truly captivating, demonstrating the tool's ability to transform videos in creative ways.

However, the author's personal experiences with the feature have been mixed. While the demos appear to be carefully curated, the author's own attempts have not yielded the same level of success. The results have been inconsistent, with the tool struggling to maintain the author's appearance and producing questionable audio effects.

The author acknowledges that there may be better prompting practices or techniques that they are not yet aware of, which could potentially yield better results. Nevertheless, the discrepancy between the polished demos and the author's own experiences highlights the need for further refinement and improvement in the tool's performance.

Overall, Luma AI's modify video feature holds great promise, but the author's experiences suggest that the technology is still a work in progress. As the field of AI-powered video transformation continues to evolve, the author remains hopeful that Luma and other similar tools will continue to improve and deliver more consistent and reliable results in the future.

The past week has seen significant advancements in the realm of AI-powered video and avatar generation tools. Several companies have unveiled impressive new features and capabilities, pushing the boundaries of what's possible with these technologies.

One of the standout developments is Luma AI's "Modify Video" feature. This tool allows users to upload their own videos and then reimagine them in different styles, similar to what was seen with Runway's Gen 1 technology, but with significantly more impressive results. The most intriguing aspect is the ability to seamlessly change the outfits and environments of characters within the video, while maintaining their consistent appearance.

Another notable advancement comes from Hey Genen, which has upgraded its Avatar 4 to make AI avatars look and sound even more realistic. The demo videos showcase impressive lip-syncing and expressive movements, though some users may still detect an uncanny quality to the voices.

Captions, another player in this space, has also demonstrated its ability to create AI avatars with highly realistic voices and lip-syncing. The examples provided suggest a slightly more emotive and natural-sounding performance compared to Hey Genen's offerings.

Higsfield AI, a relatively new player in the market, has also added lip-syncing capabilities to its video tools, allowing users to create characters that appear to speak directly to the camera. While the overall quality may not be as polished as some of the other solutions, Higsfield's rapid expansion and feature updates have made it a company to watch in the AI video generation landscape.

Lastly, Menace's new video generation tool has also caught attention, with its demo videos showcasing impressive results that appear to be on par with the best offerings in the market. As with many of these tools, the true test will be in how well they perform with user-generated content beyond the carefully curated demos.

Overall, the advancements in AI avatar and video generation tools are truly remarkable, and the continued progress in this field is sure to have a significant impact on content creation, virtual communication, and beyond.

If you're an AI enthusiast, creator, or someone who collaborates on tech projects, you probably know the pain of juggling multiple tools for meetings, messaging, and file sharing. Microsoft has made it insanely easy to streamline everything, and it's free right now.

AI is moving at breakneck speed, and collaboration is key. Whether you're testing new AI tools, brainstorming startup ideas, or discussing the latest AI advancements, you probably bounce between five different apps - one for meetings, another for chat, and another for file sharing. It's clunky, inefficient, and it slows you down.

But Microsoft Teams Free gives you everything in one app: video calls, unlimited chats, shared docs, and even a community feature that's perfect for AI masterminds and startup teams. It's built for early adopters like us, people who thrive on innovation and need tools that keep up.

Microsoft Teams Free includes 60 minutes of free video meetings, unlimited chat with no message deletion, 5 GB of free OneDrive storage for easy file sharing, and it works across desktop, mobile, and web seamlessly across devices. It's perfect if you're running a side hustle, launching a startup, or just working with a group of AI-curious friends who want to stay connected.

I've been testing it out, and honestly, it just makes sense. No more app switching, no more chaos - just everything you need for free. Check it out today at aka.m/futuretools if you're serious about AI tech or just getting things done. This is a no-brainer.

Phonely, a company that specializes in AI-powered phone interactions, has developed an AI agent that has achieved a remarkable 99% accuracy in customer calls. By partnering with Matai and Grock, the chipmaker, Phonely has been able to significantly improve the response time of their AI, making it sound more natural and realistic to the people on the other end of the call.

The company claims that customers are now unable to distinguish the AI agent from a human representative. This advancement in AI-powered customer service has both exciting and concerning implications. On one hand, it could streamline processes and provide more efficient assistance to customers. However, there are also concerns about the potential for abuse, such as the use of this technology for scams.

While the technology is impressive, it raises questions about transparency and the ethical implications of AI-powered interactions that are indistinguishable from human-to-human communication. As this technology continues to evolve, it will be important to consider the balance between innovation and the potential risks to consumers and society.

11 Labs also rolled out a new voice feature this week with 11 Labs version 3. It can even do like conversations back and forth with people and you can add in things like chuckles and laughing and excited and shouting. Here's one of their dialogue examples:

Hey Chris, knock-knock. I'm not doing this again, come on man, please. I promise you'll love this one.

The feature sounds really, really good. The author went ahead and jumped over to 11 Labs and trained their own voice into it. When you train your voice, they really give you two options - a quick option where they just need like 30 seconds of audio, and a better quality option where you train on like 30 minutes of audio. The author just did the 30 seconds of audio, so the result may not sound as good, but here's what it made from their voice:

Everyone should subscribe to Matt Wolf and check out Future Tools now!

The author notes that in their own head, it doesn't really sound like them, but they're sure people in the comments will say it does sound just like them. They have this weird cognitive dissonance when it comes to hearing their own voice.

We also got an update out of Sunno, the AI music platform. Most people already know that Sunno is probably one of the best, if not the best, music generation platforms out there that can generate really high quality music. But now, you can even do things like reorder, rewrite, and remake your tracks section by section. You can extract specific stems like just the vocals, just the drums, just the bass or guitar, piano, etc. and pull those tracks and then pull them into your own audio editor and sort of remix them. They've also upgraded the creative control that you get when actually using Sunno.

This week, OpenAI rolled out a few notable updates to ChatGPT:

1. Memory Feature for Free Users: The memory feature, previously only available for paid plans, is now accessible to free users as well. This feature allows ChatGPT to learn from past conversations and use that context to provide more personalized and relevant responses. Users can now ask questions like "Based on everything you know about me, what should I focus on in my life right now?" and get tailored advice.

2. Connectors for Paid Plans: OpenAI has introduced the ability to connect ChatGPT to various productivity tools, such as Outlook, Microsoft Teams, Google Drive, Gmail, SharePoint, Dropbox, and Box. This integration allows users on the Plus and Pro plans to have ChatGPT pull information from these sources during conversations, making it a more powerful and versatile tool for tasks like research, analysis, and content creation.

These updates further enhance the capabilities of ChatGPT, making it more personalized and integrated with the tools users rely on daily. The memory feature for free users is a particularly notable addition, as it brings more advanced functionality to the base version of the AI assistant. The connectors for paid plans also demonstrate OpenAI's efforts to make ChatGPT a more comprehensive productivity tool, blending AI-powered language abilities with access to users' own data and files.

There's been some drama in the AI world this week. First, it seems that there is tension between OpenAI and Anthropic.

Verun Mohan, the CEO of Windsurf (a company that OpenAI is reportedly acquiring), shared that Anthropic has cut off nearly all of Windsurf's access to Anthropic's Claude 3.x models, with less than 5 days' notice. Windsurf was disappointed by this decision and the short notice, and it seems they never got a clear answer from Anthropic on why this happened.

Adding more context, the CEO of Anthropic, Dario Amodei, used to work at OpenAI. This suggests there may be some underlying tensions or competition between the two AI companies.

In another drama, Reddit is suing Anthropic. Apparently, Anthropic's bots have accessed Reddit's data more than 100,000 times since last year. Reddit has a deal with Google to allow Google to use Reddit's data for its own AI training, but they don't have a similar deal with Anthropic. Reddit is not happy that Anthropic is apparently bypassing their APIs and scraping information to train Anthropic's models.

It's an interesting web of connections, as Google is actually a 14% owner of Anthropic. The relationships and tensions between these AI companies and their data usage practices seem to be a growing area of drama and legal battles in the industry.

• Apparently, the drama surrounding Sam Altman's firing and rehiring at OpenAI is being turned into a movie called "Artificial" directed by Luca Guadagnino. The film will be produced for Amazon MGM.

• Microsoft Bing has made Sora, the AI video generation tool, available for free, but with the catch that you have to use it within the Bing app.

• Google has updated its Gemini 2.5 Pro model, which has shown improvements in text generation and outperformed Anthropic's Claude Opus 4 in code generation, according to benchmarks.

• The content creation tool Opus Clip has rolled out a new feature called Opus Search, which tracks trends and alerts creators when they can repurpose their past content into short-form videos.

• In a somewhat dystopian move, Meta is planning to replace humans with AI to assess privacy and societal risks on its platforms.

• In a more lighthearted note, the ASMR YouTuber Yeti Boo has shared some new quiet, soothing content.

The story of Builder AI's deception highlights the importance of scrutinizing claims made by AI companies, especially those seeking significant investments. While the AI industry continues to evolve rapidly, it is crucial to maintain a critical eye and ensure that the technology is being developed and deployed responsibly.

The emergence of various video transformation tools, such as Luma AI's modify feature and the advancements in AI-generated avatars, demonstrates the ongoing progress in this domain. However, as the examples showcased, the results can still be inconsistent, underscoring the need for further refinement and best practices in prompt engineering.

The developments in audio generation, with companies like Phonely and 11 Labs pushing the boundaries of realism, raise important ethical considerations. While these technologies may offer convenience, the potential for misuse, such as scams, must be carefully addressed.

The drama unfolding between OpenAI, Anthropic, and Reddit serves as a reminder of the complex web of relationships and power dynamics within the AI ecosystem. These tensions underscore the need for greater transparency and accountability in the industry.

As the AI landscape continues to evolve, it is essential to remain vigilant, critically evaluate claims, and prioritize responsible development and deployment of these transformative technologies. The future of AI holds immense potential, but it must be navigated with care and foresight to ensure it benefits society as a whole.