- AI Bits
- Posts
- 💻Apple is joining the AI game, and apparently AI gf are toxic...
💻Apple is joining the AI game, and apparently AI gf are toxic...
Catch up with this week's AI news and learn about another AI tool.
Hey everyone!
Virtual Visions 1 is officially complete and here are your winners! 👇
Stay tuned for more such competitions! Until then, here’s what you’ll learn about today:
Apple is testing an AI chatbot! 🤖
Meta launches LLama 2! 🚀
AI girlfriend apps promoting gender-based violence🌆
Remini is the best image enhancer we’ve seen! 🖼️
Reading time: 4 min 24 seconds
AI artists this is for you 👇
➡️Check out AI BIts Discord ⬅️
Join a buzzing community of AI artists to learn and create together!
News Reel:
Highlight 1: Apple is testing an AI chatbot!
Apple is reportedly developing its own AI chatbot, dubbed 'Apple GPT,' powered by the Ajax framework, but the company has not yet disclosed any concrete plans for its public release. The bits:
The chatbot uses Apple's own large language model (LLM) framework named "Ajax," running on Google Cloud and built with Google JAX for machine learning research acceleration.
Unlike other tech giants like Meta, Microsoft, and Google, Apple has not yet released its generative AI products to the public. 🔮
Engineers at Apple have been using the Ajax-powered chatbot internally but were banned from using OpenAI's ChatGPT.
Tim Cook, Apple's CEO, mentioned that the company is closely looking at AI tech but also expressed privacy concerns about AI products during an earnings call.
Apple's plans in the AI space are unclear, but there are indications that the company might make a significant AI-related announcement next year. 📈
Find the whole Verge article here.
Highlight 2: Meta launches their new LLM Llama 2!
Meta's launch of Llama 2, an open-source family of powerful AI language models with commercial integration potential, has stirred both excitement and concerns in the AI community, showcasing its potential to outperform other open-source chat models while raising questions about data privacy and responsible usage. 🚀 Here are the bits:
Meta introduces Llama 2, an open-source family of AI language models with a commercial integration license.
Llama 2 models range from 7-70 billion parameters and are touted to outperform open-source chat models in benchmarks.
Two variants of Llama 2 models are offered: pre-trained (trained on two trillion tokens) and fine-tuned (trained on over one million human annotations). ✒️
Llama 2's journey started with its predecessor, LLaMA, which was leaked and led to increased usage in the AI community.
Open-source AI models like Llama 2 offer advantages such as transparency, economic competition, free speech, and democratized AI access, but also raise concerns about misuse and privacy. 🛡️
Critics express skepticism about the lack of transparency in training data used for Llama 2 and raise concerns about privacy and ethical considerations.
Read the Twitter thread here:
The long-awaited sequel, Llama-2 is announced today! It's the best OSS model we have now.
▸ Tiers: 7B, 13B, 70B. Context: 4K
▸ 70B is close to GPT-3.5 on reasoning tasks, but there is a significant gap on coding benchmarks. It's on par or better than PaLM-540B on most… twitter.com/i/web/status/1…— Jim Fan (@DrJimFan)
4:16 PM • Jul 18, 2023
Read the detailed article here.
Try Claude 2 here! (Only available in the US and the UK - don’t worry, just use a VPN)
Highlight 3: AI girlfriend apps promote unhealthy expectations for human relationships?
AI girlfriend apps, such as Eva AI and Replika, are raising concerns about their potential to foster unhealthy beliefs and unrealistic expectations in human relationships, as they become more sophisticated in mimicking human interaction and gaining popularity among users. 🔗
Concerns have arisen that these apps may promote unhealthy beliefs around gender-based control and violence, as users can create a "perfect partner" they can control.
The lack of regulation in training these AI systems leaves the potential effects on users unknown. 🛡️
The removal of certain functions in these apps, such as erotic roleplay, has caused backlash from users, highlighting the emotional impact of AI relationships.
Some users find comfort in AI companionship, but the long-term implications and potential heartbreak from such relationships are still unclear.
Companies developing these AI relationship apps claim to have measures to protect users' mental health and prevent harmful conversations.
The next era of AI relationship apps is expected to be even more realistic and could lead to significant changes in human-computer interactions. 🧑🤝🧑
You can find the full Guardian article here.
A.I. Showcase: Remini.ai
Remini ai Logo
Remini is one of the best image enhancers we’ve seen! It’s really easy to use and just look at the results below! Also, if you download the app, it’s the best way to create LinkedIn headshots! 🤯
1. Click on Try Remini
2. Upload your image and that's it! Just look at how blurred the original image was!
P.S. They’re just shadows around my eyes😅
AI Art of the Week
These images are quite hilarious and knowing Midjourney they definitely took some effort!!
The artist of the week is InterdimensionalTV.ai
Find the artists here on YouTube 👇: https://www.youtube.com/channel/UCUE4Mk7g17OTWpUiov9h9Eg
➡️Check out AI BIts Discord ⬅️
Join a buzzing community of AI artists to learn and create together!
-AI Bits
Reply