All about AI, Web 3.0, BCI
3.71K subscribers
769 photos
28 videos
162 files
3.46K links
This channel about AI, Web 3.0 and brain computer interface(BCI)

owner @Aniaslanyan
Download Telegram
OpenAI now projects $125B in revenue in 2029, with $25B of that from new products not yet announced

The Information reports OpenAI forecasts revenue reaching $125 billion in 2029 and $174 billion in 2030, mainly from AI agents, subscriptions, monetizing free users, and potentially affiliate fees.

According to internal documents seen by The Information, OpenAI expects revenue in 2029 to include $29 billion from AI agents, $50 billion from ChatGPT subscriptions, $22 billion from API access, and $25 billion from monetizing free users and other new, unspecified products.

CEO Sam Altman mentioned recently affiliate fees or taking a percentage of sales generated through user searches as possible revenue sources, while CFO Sarah Friar told the Financial Times there are โ€œno active plansโ€ for selling traditional advertising.

If they hit it, the current valuation ($300B) will be a steal; Google does ~$400B in revenue and is worth $2T.
๐Ÿ‘3โค2๐Ÿ‘2
Trends in AI Supercomputers

Epoch AI dropped a map of the worldโ€™s 500+ AI supercomputers.

โ€ข Performance doubling every 9 months.
โ€ข Hardware cost & power use doubling every year.
โ€ข xAIโ€™s Colossus already gulps 300 MW โ€” equal to 250 k homes โ€” and thatโ€™s only 2025.

If the trendline holds, the 2030 front-runner will burn 9 GW, pack 2 M chips, and sport a $200 B price tag.

The U.S. owns 75 % of todayโ€™s compute muscle while China trails at 15 %.
๐Ÿ”ฅ7โค3๐Ÿ‘2
MIT researchers have developed a "periodic table" for machine learning โ€” a groundbreaking framework that maps the connections between 20+ classical ML algorithms.

By revealing how these methods relate and overlap, the table opens up new ways for scientists to hybridize techniques, improving existing models or even inventing entirely new ones.

As proof of concept, the team fused two distinct algorithms using this framework and created a novel image classification method โ€” outperforming current state of the art models by 8%.
๐Ÿ”ฅ5๐Ÿ†’5โค3๐Ÿ‘1
Itโ€™s announcement about the new lightweight ChatGPT deep research model (powered by a version of o4-mini) and updated limits confusing

In the end, it's a combination of tasks using "standard" deep research plus additional tasks using the lightweight version:

- Free - 5 tasks/month using the lightweight version
- Plus & Team - 10 tasks/month, plus an additional 15 tasks/month using the lightweight version
- Pro - 125 tasks/month, plus an additional 125 tasks/month using the lightweight version
- Enterprise - 10 tasks/month
๐Ÿ”ฅ3โค2๐Ÿ‘2
Liquid AI introduced architecture called Hyena Edge, a convolution-based multi-hybrid model that not only matches but outperforms strong Transformer-based baselines in computational efficiency and model quality on edge hardware, benchmarked on the Samsung S24 Ultra smartphone.

To design Hyena Edge, researchers used end-to-end automated model design framework โ€”STAR.
โค3๐Ÿ‘3๐Ÿ‘2
Why Do Multi-Agent LLM Systems Fail? Despite the growing excitement around Multi-Agent Systems (MAS), they often struggle to outperform single-agent approaches.

Berkeleyโ€™s researchers analyzed 7 popular MAS frameworks across 200+ tasks, identifying 14 failure modes that hinder their effectiveness.

Paper.
Code.
โค5๐Ÿ‘3๐Ÿ‘2
Stripe is building a NEW stablecoin product, powered by Bridge

If your company is:
- Based outside of the US, EU, or UK
- Interested in dollar access

Send a quick note about your company to stablecoins@stripe.com
๐Ÿ”ฅ5๐Ÿ’ฏ3๐Ÿคฉ2๐Ÿ‘1
Perplexity released an agentic Voice Assistant

It uses web browsing and multi-app actions to book reservations, send emails and calendar invites, play podcasts/videos, and more

Currently available in the Perplexity app, but only on iOS
โšก๏ธViral rumors of DeepSeek R2 leaked

โ€”1.2T param, 78B active, hybrid MoE
โ€”97.3% cheaper than GPT 4o ($0.07/M in, $0.27/M out)
โ€”5.2PB training data. 89.7% on C-Eval2.0
โ€”Better vision. 92.4% on COCO
โ€”82% utilization in Huawei Ascend 910B

Big shift away from US supply chain.
A new paper from Google DeepMind shows how Reinforcement Learning Fine-Tuning (RLFT) on self-generated Chain-of-Thought (CoT) can improve exploration and decision-making.

RLFT Implementation:
1. Set up the LLM to interact with a decision environment (e.g., bandit, Tic-Tac-Toe).
2. Prompt the LLM to generate a thinking process (CoT) and an action.
3. Extract and execute the action in the environment to get a reward.
4. Use the reward to fine-tune the LLM (via RL) based on the generated CoT and action.
5. Repeat interactions and fine-tuning to improve the LLM's decision policy.

Insights:
-LLMs act greedily, sticking to early successful actions and failing to explore potentially better options.
-  Smaller LLMs tend to repeat actions common in the prompt history.
- LLMs can often articulate or calculate the correct strategy, but fail to execute
- Simple Reward bonuses (e.g., +1 for exploring a new action) or penalties (e.g., -5 for an invalid action format) guide LLM towards desired behaviour.
- Thinking Time Matters, allowing more tokens for generation improves performance
- Larger models suffer less frequency bias but are still prone to greediness
๐Ÿ‘8โค4๐Ÿ‘2
HuggingFace introduced SO-101 the new version of the hugely popular SO-100 low-cost robot arm:
- easier to assemble
- more robust in daily use
- still 100% open-source
- still ultra low-cost

Wowrobot shop
Seeedstudio shop
Partabot shop
๐Ÿ”ฅ7โค3๐Ÿ‘2
Mastercard announced the launch of a global stablecoin payment system, covering wallet enablement, card issuance, merchant settlement, and on-chain remittances, and will partner with OKX to issue the OKX Card, linking crypto trading with everyday spending.

Mastercard is also collaborating with Circle, Nuvei, and Paxos to enable direct merchant settlement in stablecoins.
๐Ÿ”ฅ4โค3๐Ÿ‘2
OpenAI is introducing shopping features in ChatGPT today, powered by GPT-4o, for ChatGPT Pro, Plus, and Free users, as well as logged-out users worldwide.
๐Ÿ”ฅ4โค3๐Ÿ‘2
Alibaba Introduced Qwen3

Open-weight Qwen3, latest large language models, including 2 MoE models and 6 dense models, ranging from 0.6B to 235B.


Qwen3-235B-A22B, achieves competitive results in benchmark evaluations of coding, math, general capabilities, etc., when compared to other top-tier models such as DeepSeek-R1, o1, o3-mini, Grok-3, and Gemini-2.5-Pro.

Additionally, the small MoE model, Qwen3-30B-A3B, outcompetes QwQ-32B with 10 times of activated parameters, and even a tiny model like Qwen3-4B can rival the performance of Qwen2.5-72B-Instruct.

Trained on 36T tokens, covering 119 languages! Data extracted from PDFs, synthetic data, etc.
Thinking and non-thinking modes
Improved agentic, coding capabilities, support for MCP
Training pipeline similar to DeepSeek R1
Small distilled models, such as Qwen3-4B that can rival the performance of Qwen2.5-72B-Instruct, even a Qwen3-0.6B model

GitHub
HuggingFace
Modelscope
๐Ÿ‘4โค3๐Ÿ‘2
New work on automated prompt engineering for personalized text-to-image generation:

PRISM: Automated Black-box Prompt Engineering for Personalized Text-to-Image Generation

Paper + Code

Prompt engineering for personalized image generation is labor-intensive or requires model-specific tuning, limiting generalization.

Key Idea: PRISM uses VLMs and iterative in-context learning to automatically generate effective, human-readable prompts using only black-box access to image generation models.

This approach shows strong generalization and versatility in generating accurate prompts for objects, styles and images across multiple T2I models, including Stable Diffusion, DALL-E, and Midjourney. It also enables easy editing and multi-concept prompt generation.
BCG_AI_Agents_MCP_1745919815.pdf
22.8 MB
BCG ๐—ฑ๐—ฟ๐—ผ๐—ฝ๐—ฝ๐—ฒ๐—ฑ ๐˜๐—ต๐—ฒ๐—ถ๐—ฟ ๐—น๐—ฎ๐˜๐—ฒ๐˜€๐˜ ๐—ฃ๐—ข๐—ฉ ๐—ผ๐—ป ๐—”๐—œ ๐—”๐—ด๐—ฒ๐—ป๐˜๐˜€ ๐—ฎ๐—ป๐—ฑ ๐˜๐—ต๐—ฒ ๐— ๐—ผ๐—ฑ๐—ฒ๐—น ๐—–๐—ผ๐—ป๐˜๐—ฒ๐˜…๐˜ ๐—ฃ๐—ฟ๐—ผ๐˜๐—ผ๐—ฐ๐—ผ๐—น (๐— ๐—–๐—ฃ)

๐—›๐—ฒ๐—ฟ๐—ฒ ๐—ฎ๐—ฟ๐—ฒ ๐—ธ๐—ฒ๐˜† ๐˜๐—ฎ๐—ธ๐—ฒ๐—ฎ๐˜„๐—ฎ๐˜†๐˜€:

1. ๐—”๐˜‚๐˜๐—ผ๐—ป๐—ผ๐—บ๐—ผ๐˜‚๐˜€ ๐—”๐—ด๐—ฒ๐—ป๐˜๐˜€ ๐—”๐—ฟ๐—ฒ ๐— ๐—ผ๐˜ƒ๐—ถ๐—ป๐—ด ๐—™๐—ฟ๐—ผ๐—บ ๐—–๐—ผ๐—ป๐—ฐ๐—ฒ๐—ฝ๐˜ ๐˜๐—ผ ๐—ฅ๐—ฒ๐—ฎ๐—น๐—ถ๐˜๐˜†:
โžœ Early deployments are already delivering 30โ€“90% improvements in speed, productivity, and cost across coding, compliance, and supply chain domains.

2. ๐— ๐—–๐—ฃ ๐—œ๐˜€ ๐—•๐—ฒ๐—ฐ๐—ผ๐—บ๐—ถ๐—ป๐—ด ๐˜๐—ต๐—ฒ ๐—•๐—ฎ๐—ฐ๐—ธ๐—ฏ๐—ผ๐—ป๐—ฒ ๐—ผ๐—ณ ๐—ฆ๐—ฐ๐—ฎ๐—น๐—ฎ๐—ฏ๐—น๐—ฒ ๐—”๐—ด๐—ฒ๐—ป๐˜๐˜€:
โžœ The Model Context Protocol (MCP) is the new open standard adopted by Anthropic, OpenAI, Microsoft, Google, and Amazon to expose tools, prompts, and resources reliably.

3. ๐—”๐—ด๐—ฒ๐—ป๐˜ ๐—œ๐—ป๐˜๐—ฒ๐—น๐—น๐—ถ๐—ด๐—ฒ๐—ป๐—ฐ๐—ฒ ๐—œ๐˜€ ๐—ฃ๐—ฟ๐—ผ๐—ด๐—ฟ๐—ฒ๐˜€๐˜€๐—ถ๐—ป๐—ด ๐—ฅ๐—ฎ๐—ฝ๐—ถ๐—ฑ๐—น๐˜†:
โžœ Agents today can automate tasks up to one hour long โ€” and this limit is doubling every seven months, pushing toward multi-day autonomous workflows by the end of the decade.

4. ๐—”๐—ด๐—ฒ๐—ป๐˜ ๐—”๐—ฟ๐—ฐ๐—ต๐—ถ๐˜๐—ฒ๐—ฐ๐˜๐˜‚๐—ฟ๐—ฒ๐˜€ ๐— ๐˜‚๐˜€๐˜ ๐—•๐—ฒ ๐—ฆ๐—ฒ๐—ฐ๐˜‚๐—ฟ๐—ถ๐˜๐˜†-๐—™๐—ถ๐—ฟ๐˜€๐˜:
โžœ Security challenges grow as agents gain system access. OAuth, RBAC, permission isolation, eval-driven development, and real-time monitoring are mandatory to deploy agents safely.

5. ๐—ง๐—ต๐—ฒ ๐—ฅ๐—ถ๐˜€๐—ฒ ๐—ผ๐—ณ ๐—”๐—ด๐—ฒ๐—ป๐˜-๐—ข๐—ฟ๐—ฐ๐—ต๐—ฒ๐˜€๐˜๐—ฟ๐—ฎ๐˜๐—ถ๐—ผ๐—ป ๐—ฃ๐—น๐—ฎ๐˜๐—ณ๐—ผ๐—ฟ๐—บ๐˜€:
โžœ Platforms like Azure Foundry, Vertex AI, Bedrock Agents, and Lindy are positioning themselves as the orchestration layer to create, manage, and scale enterprise agent ecosystems.

6. ๐—™๐—ฟ๐—ผ๐—บ ๐—ช๐—ผ๐—ฟ๐—ธ๐—ณ๐—น๐—ผ๐˜„๐˜€ ๐˜๐—ผ ๐—™๐˜‚๐—น๐—น๐˜† ๐—”๐˜‚๐˜๐—ผ๐—ป๐—ผ๐—บ๐—ผ๐˜‚๐˜€ ๐—”๐—ด๐—ฒ๐—ป๐˜๐˜€:
โžœ Enterprises are shifting from prompt chaining (rigid workflows) to fully autonomous agents capable of observing, reasoning, and acting dynamically based on real-world feedback.

7. ๐— ๐—–๐—ฃ ๐—ฎ๐—ป๐—ฑ ๐—”2๐—” ๐—ช๐—ถ๐—น๐—น ๐——๐—ฒ๐—ณ๐—ถ๐—ป๐—ฒ ๐˜๐—ต๐—ฒ ๐—”๐—ด๐—ฒ๐—ป๐˜ ๐—˜๐—ฐ๐—ผ๐—ป๐—ผ๐—บ๐˜†:
โžœ MCP connects agents to tools and data. A2A (Agent-to-Agent communication) will enable agents to negotiate, collaborate, and coordinate across systems โ€” forming true multi-agent networks.
๐Ÿ†’3๐Ÿ”ฅ2