* Originally in: Dovenet TechTalk

From TLDR 2024-02-14

OPENAI EXPERIMENTS WITH GIVING CHATGPT A LONG-TERM CONVERSATION
MEMORY (3 MINUTE READ) [5]

OpenAI is experimenting with adding a form of long-term memory to
ChatGPT. Users will be able to ask the chatbot to remember something,
see what it remembers, and ask it to forget. The feature is currently
available to a small number of users for testing. OpenAI hasn't
revealed what technique it is using for the system, but the
implementation is similar to that of Custom Instructions, a feature
that was introduced in July last year.

EARLY DEMENTIA DIAGNOSIS: BLOOD PROTEINS REVEAL AT-RISK PEOPLE (3
MINUTE READ) [7]

A study has identified blood proteins that appear to predict the risk
of developing dementia up to 15 years before diagnosis. High levels of
the blood proteins GFAP, NEFL, GDF15, and LTBP2 are strongly
associated with dementia. Blood levels of these proteins can be
outside of normal ranges more than 10 years before symptom onset. The
findings, once validated, may lead to the development of blood tests
that identify people at risk of developing dementia.

SOMEONE IS GOING TO DIM THE SUN, AND IT WILL BE SOON (12 MINUTE READ) [8]

Solar Radiation Management (SRM) is starting to gain more attention
due to 2023's high temperatures. It's much easier to artificially dim
the sun than most people imagine. Injecting sulfur into the
stratosphere could be a cheap way to cool down the Earth. It would
have an almost instant effect, but once it starts it would be
dangerous to suddenly stop. SRM will not fix the root cause of the
problem, but it could buy time for us to remove CO2 from the
atmosphere. However, its implementation could make us more complacent
about fixing the core issues.

META'S AI CHIEF YANN LECUN ON AGI, OPEN-SOURCE, AND AI RISK (11
MINUTE READ) [12]

Yann LeCun, Meta's chief AI scientist, received a TIME100 Impact
Award for his contributions to the world of artificial intelligence on
Sunday. LeCun sat down with TIME before the ceremony to discuss
barriers to achieving artificial general intelligence, the merits of
Meta's open-source approach, and claims that AI could pose an
existential risk to the human race. This article contains an edited
transcript of the interview.

'I'VE NEVER SEEN IT THIS BAD:' GAME DEVELOPERS EXPLAIN THE HUGE
LAYOFFS HITTING RIOT, EPIC, AND MORE (19 MINUTE READ) [15]

The games industry is undergoing a trend of mass layoffs that seems
poised to span at least two years.

MOZILLA DOWNSIZES AS IT REFOCUSES ON FIREFOX AND AI (5 MINUTE READ) [17]

Mozilla plans to shut down Hubs and scale back its investment in its
VPN, Relay, and Online Footprint Scrubber.

APPLE WON'T BE FORCED TO OPEN IMESSAGE TO RIVALS, EU DECIDES, AS IT
ALSO LETS THREE MICROSOFT SERVICES OFF DMA HOOK (4 MINUTE READ) [18]

Microsoft no longer faces tighter controls on how it can operate Bing
in the EU - the EU commission has also closed investigations into
Microsoft's Edge and online advertising business.

Links:
------
[5] https://arstechnica.com/information-technology/2024/02/amnesia-begone-soon-chatgpt-will-remember-what-you-tell-it-between-sessions/?utm_source=tldrnewsletter

[7]
ttps://www.nature.com/articles/d41586-024-00418-9?utm_source=tldrnewsletter

[8] https://climate.benjames.io/someone-is-going-to-dim-the-sun/?utm_source=tldrnewsletter

[12] https://time.com/6694432/yann-lecun-meta-ai-interview/?utm_source=tldrnewsletter

[15] https://www.ign.com/articles/ive-never-seen-it-this-bad-game-developers-explain-the-huge-layoffs-hitting-riot-epic-and-more?utm_source=tldrnewsletter

[17] https://techcrunch.com/2024/02/13/mozilla-downsizes-as-it-refocuses-on-firefox-and-ai-read-the-memo/?utm_source=tldrnewsletter

[18] https://techcrunch.com/2024/02/13/apple-wont-be-forced-to-open-imessage-to-rivals-eu-decides-as-it-also-lets-three-microsoft-services-off-dma-hook/?utm_source=tldrnewsletter

-- https://tldr.tech/tech?ref=6061602

$$
--- SBBSecho 3.14-Linux
* Origin: capitolcityonline.net * Telnet/SSH:2022/HTTP (1:2320/105)