Lifeboat Foundation

An international research group has for the first time reconstructed ancestors dating back 2.6 billion years of the well-known CRISPR-Cas system, and studied their evolution over time. The results suggest that the revitalized systems not only work, but are more versatile than current versions and could have revolutionary applications. Nature Microbiology has published the results of this research, which, in the opinion of the research team, “opens up new avenues for gene editing.”

The project, led by Ikerbasque research professor Rául Pérez-Jiménez of CIC nanoGUNE, involves teams from the Spanish National Research Council, the University of Alicante, the Rare Diseases Networking Biomedical Research Center (CIBERER), the Ramón y Cajal Hospital-IRYCIS and other national and international institutions.

The acronym CRISPR refers to the repeated sequences present in the DNA of bacteria and archaea (prokaryotic organisms). Among the repeats, these microorganisms harbor fragments of genetic material from viruses that infected their ancestors; that enables them to recognize a repeat infection and defend themselves by cutting the invaders’ DNA using Cas proteins associated with these repeats. It is a mechanism (CRISPR-Cas system) of antiviral defense. This ability to recognize DNA sequences is the basis of their usefulness, and they act as if they were molecular scissors. Nowadays CRISPR-Cas technology enables pieces of genetic material to be cut and pasted into any cell, so that it can be used to edit DNA.

ChatGPT could make Bing search results more humanlike.

The Memo: https://lifearchitect.ai/memo/

Demo site: https://muse-model.github.io/
Read the paper: https://arxiv.org/abs/2301.

Continue reading “First look — Muse by Google AI/Research — Launched 2/Jan/2023 — (3B + 4.6B T5-XXL) — Google Muse” »

Proving you’re a human on a web flooded with generative AI content.

Nadella highlighted that while generative AI tools, such as ChatGPT and Dall-E, generated less than 1% of the world’s AI data sets in 2021, this can increase to 10% of all data generated by AI by 2025.

“In future, the generative models will generate most of the data. We are right now seeing the emergence of a new reasoning engine. We’ll clearly have to talk about this reasoning engine — what are its responsible uses, what displacements will it cause, and so on. But on the other side, we should also think about how it can augment us in what we are doing today since it can have a huge impact on our future,” Nadella said.

“Ultimately, these tools will accelerate creativity, ingenuity and productivity across a range of tasks. It is going to be a golden age — the computer revolution created mass consumer behaviour change and productivity for knowledge workers. But, what if we could spread that productivity more evenly? To me, that is one of the biggest things to look forward to, and the way to achieve this is by building a robust data infrastructure,” he added.

In a recent study published in the journal Cureus, researchers conducted a literature review to determine whether platelet-rich plasma therapy could help slow disease progression through lung regeneration in patients with chronic respiratory disease.

Chronic respiratory disease comprises many disorders related to the respiratory system, such as asthma, chronic obstructive pulmonary disease (COPD), and interstitial lung fibrosis, of which COPD is the most fatal, causing close to 3.23 million deaths worldwide in 2019. Statistics suggest that approximately 37 million people in the United States live with chronic respiratory disease, and over 150,000 individuals in the country succumbed to the disease in 2020.

Chronic bronchitis and emphysema as the two prominent disorders constituting COPD, causing abnormal airflow into and out of the lungs. Emphysema causes the destruction of the alveoli, which is where the exchange of gases occurs in the lungs, while chronic bronchitis causes excessive mucus production and obstruction of airways.

Scientists have developed a synthetic tissue that repairs injuries and restores normal erectile function in a pig model, in a study publishing January 4 in the journal Matter. The findings suggest that the artificial tunica albuginea (ATA), which mimics a fibrous sheath of tissue necessary to maintain erections, shows promise for repairing penile injuries in humans.

“We largely foresaw the problems and results of the ATA construction process, but we were still surprised by the results in the animal experiments, where the penis regained normal erection immediately after the use of ATA,” said Xuetao Shi, a researcher at the South China University of Technology in Guangzhou, China, and an author of the study.

Continue reading “An artificial tissue restores erectile function in pigs” »

In this article, I will explain how I used GPT-3 to integrate AI into a simple productivity application. Before I go into the code and methods used, let’s have look at the final result:

I really don’t care about IQ tests; ChatGPT does not perform at a human level. I’ve spent hours with it. Sometimes it does come off like a human with an IQ of about 83, all concentrated in verbal skills. Sometimes it sounds like a human with a much higher IQ than that (and a bunch of naive prejudices). But if you take it out of its comfort zone and try to get it to think, it sounds more like a human with profound brain damage. You can take it step by step through a chain of simple inferences, and still have it give an obviously wrong, pattern-matched answer at the end. I wish I’d saved what it told me about cooking and neutrons. Let’s just say it became clear that it did was not using an actual model of the physical world to generate its answers.

Other examples are cherry picked. Having prompted DALL-E and Stable Diffusion quite a bit, I’m pretty convinced those drawings are heavily cherry picked; normally you get a few that match your prompt, plus a bunch of stuff that doesn’t really meet the specs, not to mention a bit of eldritch horror. That doesn’t happen if you ask a human to draw something, not even if it’s a small child. And you don’t have to iterate on the prompt so much with a human, either.

Competitive coding is a cherry-picked problem, as easy as a coding challenge gets… the tasks are tightly bounded, described in terms that almost amount to code themselves, and come with comprehensive test cases. On the other hand, “coding assistants” are out there annoying people by throwing really dumb bugs into their output (which is just close enough to right that you might miss those bugs on a quick glance and really get yourself into trouble).