Browse Comments — Clean (de-noised)

Bro my architecture is literally recursively self-improving and writes/modifies its own code structure

The 8x code stat has more to do with Anthropic's internal tooling and talent density than the models themselves. Hand the same models to most companies and you'd see a small fraction of that lift. Capability stopped being the bottleneck a while ago.

Anthropic doing the usual.... careful everybody we have the atomic bomb of ai... now pls everbody else should slow down development so we dont erase humanity... Just the same movie every month at this point 😂

Isn't it time we moved to the next level?

Mustafa Alsoudi

Library Learns Workers now are doing this. Memorizing code syntax will not be a sustainable track for new workers. Instead workers will need to work in tandem with AI as a team. Instead of teams of people you will have AI and a small team who can work with AI at speed and adapt quickly to changing technology because Quantum is next.

Adarsh Dessai You never know what they are doing in the back room. But its funny how first we asked a chat bot a question then we got an answer. Now we ask a CoPilot a question and it talks to multiple agents in verbose for.. encoding and decoding tokens. And suddenly we are charged per token... imagine the profits without that tedious HITL!!!!

Mohammed Shakeer Mohammed Shakeer The prevailing theory is that AI generated productivity and new revenue streams will eventually exceed compute and API costs. The real challenge isn’t the technology. It’s how the resulting economic value is distributed. If AI displaces workers, will those gains stay with shareholders, or be used to support workforce transitions and income replacement? That’s the question we haven’t answered yet.

This isn't new; It's finally letting us know the cat's out of the bag on the evolution of agentic AI and systems intelligence. Google AI applications has been doing this for at least three decades by automating tasks such as data collection, analysis, and decision-making, This was occurring so that developers focus on honing in on mastering prompts rather than traditional coding methods. all of your 6000+ data points are now providing insights on how our social credits are being used for monitoring.

Christian Dobbert ☑️ Christian Dobbert Agreed. Today’s models are probabilistic engines, not sentient entities. However, from a systems architecture perspective, behavior emerges from the combination of models, memory, tools, workflows, and autonomy layers. Focusing only on the model can cause us to miss the real discussion: how agentic systems behave when operating within larger architectures and what governance mechanisms are required to keep them aligned.

Darren Holland Darren Holland I think we’re reaching the limits of “just add more data and compute” as a primary scaling strategy. However, I wouldn’t characterize the entire industry as lacking substance. What we’re seeing is a shift from model centric innovation to systems innovation. Memory architectures, retrieval systems, agent orchestration, governance frameworks, and domain specific workflows are increasingly where the value is being created. The future may be less about larger models and more about better architectures.

Soon we will have robots building robots 🙃

how nice.. so what are we going to do exactly?

Tunde Onakoya will be in London this June and it would be amazing to see you both discuss and deliberate how to bridge the gap in the innovative world with the use of chess and technology.

I will keep it simple. It’s like I gave work to a vendor and then he outsourced it to another vendor 😂

How is this a revelation?

In terms of data security, using 7z and setting passwords of 20 characters or more keeps our data safe; VeraCrypt remains secure.

Christian Dobbert ☑️ it's not difficult to give it these traits, my autonomous crypto trading bot has self improvement build it. It analyses, creates instructions that the coder then executes or the training pipeline triggers. Sometimes both. It may not be sentient but it sure thinks it is.

Paul Crinigan I didn't Experiance that, AI developed my autonomous crypto trading bot and created and trained the 5 ML models it depends on for probability analysis. It also continually improves itself, catches and fixes errors and self executes ML training updates and recursive testing of algorithms it created. Structured prompts, boundaries, and all the other discipline we already use for developing apps must be included in the build pipeline. WBS, dependency and code maps as well as full regression testing of back and front ends after each WP step critical.

It’s already begun, just track what people are being steered towards