Tech News
← Home  ·  All topics

Algorithm

23 GoKawiil briefs on this topic

Spotify launches Taste Profile tool letting users edit recommendations with plain-language input

Spotify has introduced a feature called Taste Profile that allows users to type everyday phrases requesting more or less of specific artists, genres or moods. The tool is designed to give listeners direct input into how the platform's recommendation algorithm shapes their feed.

YouTube launches Custom Feeds letting users build separate recommendation lists

YouTube is rolling out Custom Feeds, a feature that lets viewers create and save multiple recommendation feeds using conversational prompts, such as one for wind-down documentaries or commute podcasts. These feeds appear on the Homepage and refresh automatically with new suggestions. The feature expands on a prior experiment and will launch this fall for US users on web, mobile, and TV.

YouTube launches Custom Feeds letting users write prompts to shape their homepage

Google unveiled Custom Feeds at its Made on YouTube event, a feature letting users type natural-language requests to generate a personalized homepage feed. The tool uses large language models to interpret prompts like genre or mood combinations and build a matching set of video suggestions. YouTube tested the feature earlier this year and plans a full rollout to all US users across mobile, desktop and TV.

YouTube launches AI-powered custom feeds using Gemini

YouTube introduced a feature called custom feeds that lets users type a prompt describing the videos they want recommended, such as a mix of video podcasts for a commute or relaxing commentary content. The feature uses Google's Gemini AI model to generate the feed, which is pinned to its own tab at the top of the YouTube homepage.

Spotify rolls out 'Taste Profile' recommendation tool to US Premium users

Spotify has launched its 'Taste Profile' feature to Premium subscribers aged 18 and up in the United States, after debuting it in beta in New Zealand earlier this year. The tool lets users see how Spotify's AI interprets their listening habits and adjust it using natural-language requests to change what gets recommended in feeds like Discover Weekly, Made for You, and Wrapped.

US State Department criticises Australia's proposed algorithm opt-out laws as censorship

US officials have publicly objected to Australia's proposed legislation that would let users opt out of algorithm-driven content recommendations, characterising the move as censorship. Separately, at the UN, President Trump downplayed calls for AI regulation, suggesting the technology be renamed "super intelligence" rather than artificial intelligence.

US embassy submission calls Australia's algorithm opt-out plan 'censorship'

The US embassy in Canberra formally objected to Australia's draft digital duty of care laws, which would require tech platforms to let users opt out of content algorithms or face fines. The submission argued vague definitions of 'harm' could enable viewpoint-based censorship, suppress independent journalists, and affect users globally, not just in Australia. Prime Minister Anthony Albanese, speaking during a US visit, defended the laws as returning control to individuals rather than governments.

AI agents design and Lean-verify new shortest-path algorithm beating published bounds

Ten Claude Opus 5.5 agents were tasked with finding a faster exact shortest-path algorithm for directed graphs with non-negative real weights and proving it correct in the Lean theorem prover. Over roughly 15 hours and 733 messages, the agents produced an algorithm called C-HD, which its creators say formally improves on previously published complexity bounds, including Dijkstra's algorithm and two recent 2025 and 2026 papers.

Essay warns algorithmic feeds on YouTube, Spotify, LinkedIn and Reddit are hijacking user attention

A new essay argues that platforms like YouTube, Spotify, LinkedIn and Reddit increasingly control what users see rather than letting them choose, citing doomscrolling recommendations, AI-generated filler music, sponsored opinions on LinkedIn, and bot-driven discussions on Reddit. The piece uses the 'Tetris effect' to illustrate how sustained, algorithm-driven focus reshapes thought patterns without users' explicit consent.

Cloudflare cuts server hash memory usage 90%, freeing another 100TB of RAM

Cloudflare re-tuned the hash-mapping algorithm its Pingora framework uses to route URL requests to cache servers, drastically reducing the number of hash entries needed per server. The optimization freed up roughly 100 terabytes of RAM across its infrastructure, marking a second major memory-saving effort by the company.

Blog post highlights small dev tricks like fzf history search and SQL SELECT without FROM

A blog post argues that much of engineering productivity comes from small, easily learned tricks rather than deep expertise. Examples include using python3 -m http.server for a quick local server, installing fzf for fuzzy-search of shell history, using atuin for SQLite-backed history search, and running a bare SELECT statement in SQL without a FROM clause to test expressions.

Spotify adds toggle to exclude kids' listening from recommendations and Wrapped

Spotify has introduced a new setting that lets users prevent music played for children from influencing their personalized recommendations and year-end Wrapped summary. The toggle allows adults sharing an account with kids to isolate nursery rhymes and children's music from skewing their listening data.