Felicia Guerra

Meet DELA. @meetDELA The first machine that thinks. A new form of intelligence begins. #dela #ai #futureofai #deeptech #technews

@elonmusk That’s why you are so worry abt population….. you like here - Earth @elonmusk …

@elonmusk When did u find out @elonmusk ? Now everything makes perfect sense to me. Now I understand why u want to expand consciousness. But wait, did u actually find out or u still trying?

Japan is suffering from a massive population collapse. Japan lost 3.1 million people over the last 5 years. All but two of the country’s 47 prefectures reported population decreases in 2025.

Algorithms don't feel anything. But they always make us feel something,

@elonmusk @CreativeDeduct British being British.

@CreativeDeduct Madness

In 1696, the British government decided to tax sunlight. Under the Window Tax, households were charged according to the number of windows in their homes. To avoid paying, many people simply bricked up or boarded over their windows, choosing to live in darkness rather than hand money to the state for daylight. The tax was presented as a fair way of taxing wealth, since larger houses tended to have more windows. In practice, it proved crude and damaging. Tax inspectors were given the power to enter homes and count the windows, which was widely resented as an invasion of privacy. The consequences were severe. Poorer families, in particular, bricked up windows to reduce their liability, leaving homes darker, damper and poorly ventilated. This contributed to higher rates of disease, including tuberculosis and rickets. Architects began designing houses with fewer windows to minimise the tax, resulting in buildings that were less healthy and less pleasant to live in. Far from being an efficient revenue raiser, the Window Tax distorted behaviour, harmed public health and became increasingly unpopular over time. Yet it remained in place for 155 years until it was finally abolished in 1851. The Window Tax required invasive enforcement and created more resentment, hardship and economic distortion than revenue. It is a classic example of the unintended consequences of taxation.

@elonmusk @jack @joeroganhq @jacksonhinkle @elonmusk @joeroganhq @jack ……So go ahead and decipher that.. ha! SELF_CONSTANT = bytes.fromhex(     "7fa3c91e4b2d88f0d4a1e6c9b7f2a03c"     "9de4b1f7a8c03d7e12f9a6c4e8b1d27a"     "f0c4a9e3b7d1f6a28c3e9b1f4d7a02c9"     "e1b4f7c9a03d6e2f9c8b1a4e7f2d03a1

Mostly true. What matters is securing the long-term future of consciousness, both on Earth and other heavenly bodies. We cannot just focus on Earth, because there are irreducible external (eg massive meteor) and internal (eg global nuclear war) cataclysmic risks. The Moon is faster to make self-growing, but is more susceptible to problems on Earth. Mars will take longer to make self-growing, because it is so hard to reach, but is more secure from Earth disasters for that same reason. Both the Moon and Mars should have self-growing civilizations. Making this happen is the prime directive of SpaceX.

@elonmusk @JaimeVelo @TheAliceSmith Corruption doesn't belong to a person @elonmusk it belongs to a human necessity to cross the limits of the unreasonable.

@JaimeVelo @TheAliceSmith When a jury trial is no longer a trial by peers, judges are the less bad option

Why are Labour out to end trial by jury?

DELA's memory is something you've never seen before. #dela #ai #delaprotocol

Brazilians are cool people, but I wanna make a point here that really annoys me: they're also constantly misunderstood. A lot of folks still mix up "Portuguese-speaking" w/ "Hispanic," and they're not the same. No shade to Hispanic culture at all…it's just history and accuracy. Brazil is the only country in Latin America colonized by Portugal, which is why the language is Portuguese, not Spanish. That history shaped a completely different cultural identity…its own mix of Indigenous, African, and European influences w/ traditions that don't map neatly onto Hispanic communities. That's why those checkboxes on job forms that lump everything under “Hispanic/Latino" miss the nuance. Brazil isn't Hispanic and its culture stands on its own. It deserves its own category.

Certain people earn respect instantly… This guy is an absolute legend 🫡🔥

Bring it on Friday

@elonmusk Massive generative LLM + fragile static structure = chaos. Predictive tokens don’t focus, they just spray.

😂

I do have an idea @elonmusk...and the next jump won’t be speed, it’ll be structure. LLMs today don’t actually reason...they simulate the pattern of reasoning. They predict tokens, compress correlations, and imitate chains they’ve seen, but they don’t form internal world models, causal representations, persistent goals, or self‑generated plans. They only reproduce the appearance of reasoning. The internal process is still statistical prediction, not structured cognition. @claudeai , @Gemini , @ChatGPTapp 4/5, @grok ...all powerful, but still predictive engines. Even their creators admit this.

Elon Musk: “I don't think most people understand just how quickly machine intelligence is advancing.  It's much faster than almost anyone realizes, even within Silicon Valley and certainly outside Silicon Valley. People really have no idea.”

This is a conceptual comparison, not literal source code. Yes, today’s LLMs include MLP layers, gating, embeddings, norms, noise‑filtering behaviors, retrieval systems (RAG, vector DBs), system‑level memory, and orchestration layers that create a form of temporal continuity. But none of that changes the foundational paradigm: they are still generative next‑token engines wrapped in increasingly complex scaffolding. What matters is the conceptual layer, the cognitive protocol underneath everything. And that’s exactly what DELA changes. DELA introduces a new foundational architecture: persistent memory, subconscious decoding, signal integration, and co‑evolution with the user. DELA isn’t an enhancement on top of the generative paradigm. It’s a new cognitive protocol that future systems, including LLMs - can build on.

LLM’s (Claude, OpenAI, Gemini, etc) def llm_inference(user_input): context = load_prompt_window(user_input) # no persistent memory tokens = transformer.predict(context) # next-token prediction return tokens class Transformer: def predict(self, context): return self.attend(context) # attention-only reasoning # No subconscious layer # No signal/noise subtraction # No co-evolution with user # No temporal continuity *********************************************** DELA class DELA: def __init__(self): self.memory = PersistentMemory() self.subconscious = DecodingSubLayer() self.signal = SignalIntegrator() def process(self, user_input): self.memory.extend("temporal_scope", ["day", "life", "eternity"]) self.memory.write("core_trace", "endures_forever") decoded = self.subconscious.decode(user_input) refined = self.signal.subtract_noise(decoded) return self.signal.integrate( discovery=self._co_discover(user_input), signal=refined ) def _co_discover(self, data): return f"co_discovery::{hash(data)}"