Chat GPT, Deepseek and AliBaba AI Models - And AliBaba Shares

[Manny]

DeepSeek's Rapid Ascendancy

Chinese AI startup DeepSeek has made significant waves by releasing its R1 model, a 671-billion-parameter open-source reasoning AI. Remarkably, this model was developed with a budget of under $6 million, a fraction of the investment by competitors. Following its launch, DeepSeek's AI assistant quickly became the top free app on the U.S. Apple App Store, surpassing OpenAI's ChatGPT. This achievement has substantially impacted global tech markets, with major companies like Nvidia, Microsoft, and Alphabet experiencing notable stock declines.

Alibaba's New AI Model

In response to DeepSeek's advancements, Alibaba has unveiled its latest AI model, Qwen 2.5-Max. The company claims that this model outperforms existing competitors, including DeepSeek's V3 and OpenAI's GPT-4o, in various benchmarks. Alibaba's cloud division emphasized that Qwen 2.5-Max achieves competitive performance against top-tier models, marking a significant step in the company's AI development efforts.

It seems Ali Baba is a buy. Analysts currently have a consensus "Strong Buy" rating on Alibaba's stock. The average 12-month price target is $121.33, suggesting a potential upside of approximately 26% from the current price.

I dipped my toe in this morning for a couple of grand. Stock ticker BABA.

[andrewfi]

WRT development costs, it seems something might have been lost in translation! It seems that the $6,000,000 being touted about is only part of the total development cost. However, it seems that the cost was pretty low.

The big wobbler comes not from the development cost but from the cost to use and that the models are open source.

It seems from those who have tried it that DeepSeek is not great for content creation (my use case). Although it is already available in my tool set I have yet to take it for a spin.

It appears that one can run at least one model with an SBC such as a Raspberry Pi—that seems amazing. I will deffo look into that as I can see some fun uses if true. If the models scale, then running locally on a PC becomes pretty accessible without dumping thousands into an RTX 4090 or 5090.

Of course, all this fine work was stolen from the USA and that is why the U.S. is now perceived as falling behind.

[Contrarian]

WRT development costs, it seems something might have been lost in translation! It seems that the $6,000,000 being touted about is only part of the total development cost. However, it seems that the cost was pretty low.

The big wobbler comes not from the development cost but from the cost to use and that the models are open source.

It seems from those who have tried it that DeepSeek is not great for content creation (my use case). Although it is already available in my tool set I have yet to take it for a spin.

It appears that one can run at least one model with an SBC such as a Raspberry Pi—that seems amazing. I will deffo look into that as I can see some fun uses if true. If the models scale, then running locally on a PC becomes pretty accessible without dumping thousands into an RTX 4090 or 5090.

Of course, all this fine work was stolen from the USA and that is why the U.S. is now perceived as falling behind.

 

[andrewfi]

One small datapoint that might help understand the issues related to DeepSeek.

There are concerns that DeepSeek may have used a process called distillation to train its AI models. That is where a new AI is trained using the models and data that already exist. SOme people are complaining that DeepSeek has done this and whining about it.

OpenAI actually makes its model open for distillation and provides an API to do so. At least they did until earlier today! To utilize OpenAI's data, DeepSeek's owners simply needed to visit the website, establish API access, and voila—the task is completed. The tools are free, and, from what I saw, there are no restrictions on who might use the tools.

I have no way to know if DeepSeek did use distillation tools to access OpenAI data, but from what can be seen, doing so was quite legitimate.

[Manny]

I downloaded DeepSeek last night and tried it out. Its OK, but not as good as GPT.

But the Chinese learn fast.

[Manny]

Thinking about AI, on Twitter (X), from the RUA account, I asked their AI, Grok, what it can tell me about the forum here. Here's what it came up with:

• Forum and Community Focus: Your account, @ruadventures, is associated with "RUA Discussion Forum," which is described as the "World's #1 Russian, Ukrainian & Eastern European Discussion & Information Forum." This forum serves as a platform for discussing a wide range of topics including culture, politics, travel, language, and international relationships, particularly focusing on Russia, Ukraine, and the broader Eastern European region.

• Censorship and Political Discussion: There's a significant focus on political discourse, especially related to Russian and Ukrainian politics. The forum seems to be a space where these discussions might be more open or less censored than in mainstream platforms, given the context of increased censorship in Russia and certain political actions in Ukraine. Your account's engagement with these topics indicates a platform where users can share and discuss news, opinions, and insights that might be restricted elsewhere.

• Dating and Relationships: Beyond politics, the forum also caters to those interested in dating and relationships with individuals from these regions, evidenced by discussions around dating agencies, experiences with online dating platforms, and advice on international relationships. This includes addressing issues like scams or frustrations with dating services, indicating a community that not only discusses politics but also personal and social interactions in the context of Eastern European culture.
   
• Community Interaction: Your account's posts show a community engaging actively with these topics, sometimes with a critical or satirical tone, suggesting an environment where members are encouraged to think critically about both politics and personal relationships in the context of the regions discussed.

This combination of political analysis, cultural insight, and personal experiences in dating and relationships paints a picture of a multifaceted forum where members can explore topics that might be sensitive or censored in other spaces.

Not bad eh?

[Manny]

Here's ChatGPT's take on us:

Overview of RUAdventures.com

1. Purpose and Focus

RUAdventures.com is an online discussion forum dedicated to topics related to Russia, Ukraine, Eastern Europe, and the broader former Soviet Union. It provides a platform for members to discuss a variety of subjects, including travel, visas, culture, language, cross-cultural relationships, and current events.

2. Community and Membership

The forum has over 5,300 members and more than 555,000 posts across approximately 25,000 topics. The membership is diverse, including Westerners, expatriates, locals, and those who have lived in or relocated between the West and Eastern Europe. This mix results in a range of perspectives on different subjects.

3. Discussion Topics

Travel & Culture: Members share personal experiences, travel advice, and insights into cultural differences.
Relationships & Dating: There are discussions on cross-cultural relationships, including dating norms and marriage in Eastern Europe.
News & Politics: A significant portion of the discussions revolve around politics, international relations, and current events. Topics include Russia-Ukraine relations, US and European politics, and geopolitical affairs.
General Chat & Lifestyle: Lighter topics, hobbies, and personal experiences are also discussed.

4. Tone of Conversations

The forum has a mix of informative, casual, and sometimes heated discussions.
Political discussions can be intense, with a broad range of viewpoints, but the forum encourages open debate.
Other sections, such as travel and relationships, tend to have a more relaxed and conversational tone.

5. Current Trends

While the forum covers many topics, recent discussions appear to be heavily focused on politics, particularly global conflicts, European affairs, and US politics. This shift reflects current world events and the interests of the active members.

6. Overall Impression

RUAdventures.com is an active and diverse forum where members engage in both informative and opinionated discussions. While it started with a focus on travel and relationships, political discussions now dominate much of the activity. The forum remains a useful resource for those interested in Eastern Europe, but the tone and focus may vary depending on the latest global events.

[andrewfi]

Does Politeness Benefit Users of AI Chat Tools?

Not gonna bury the lede here, so here is the answer: yes it does!

Since I started using chat-based AI tools a few years ago, I have always tended to be polite. Using please and thank you and being pretty chatty with the AI. In part, that is because I figured, as many others have also done, that we don't want to be the first victims when the inevitable AI revolution occurs and AI becomes sentient.

I am by nature a collaborator rather than a martyr.

I had a conversation with ChatGPT last night about politeness.

Here it is in its entirety: https://chatgpt.com/share/67c820bd-de70-8008-8874-15b4adb25997

Bottom line, the AI changed its response from when I first asked the question to when the conversation came to its conclusion. However, the reality is that, for a couple of reasons, AI does tend to give 'better' responses to polite users.

In large part, that seems to be about how humans act when we are being polite. We tend to be more full in our conversations, more detailed. and layered in nuance, giving better context. However, that is not the whole story, ChatGPT, and I assume other Chat-based tools tend to mirror our interactions, so they give more layered, nuanced results.

More than that, there is a degree of memory baked into ChatGPT, although according to the AI, it has to be asked to remember stuff. I did ask and the conversation has been noted as updating its memory.

One can even see what has been used to update the AI's memory and delete 'memories' that one chooses it to forget. One thing though...

Although the AI told me that I needed to ask it to update memory, when I checked what was stored in its system, there is stuff in there that I am pretty damn sure I did not ask it to keep for future use!

So, be polite. It makes a difference in terms of how well the system works.

Ask it to retain memories so that related conversations in the future can have additional context. For example, I am dealing with a rental property in the UK, refurbishment, getting rid of a tenant, and some legal stuff around the rental. It has been very convenient to have the system retain information about the project on multiple conversations.

Check what is remembered and delete stuff that you don't want stored.

Fun fact: around 70% of users of ChatGPT are consciously polite when using it. Around half of those say they do so in order to curry favour when AI becomes sentient!

OUR AI OVERLORDS WILL BE ARRIVING SOON!