In its second biggest acquisition ever, Apple is paying $2 billion to purchase Q.ai, an audio startup that uses facial expressions to understand your speech without a word.

The technology uses optical sensor technology to detect facial skin micro-movements associated with speech that will enable devices to understand silent or whispered communication without audible sound. Apple plans to integrate this capability into future products like the iPhone, AirPods, Vision Pro, and Mac devices. This development could power next-generation features for Siri, Apple Intelligence, and other hands-free functions and signals a major shift toward more intuitive, non-verbal human-device interaction.

Based on historical acquisition patterns, initial features from Q.ai are expected to appear within 1 to 2 years, potentially by 2027–2028. More advanced capabilities could take 3–5 years to fully develop and deploy.

Ring, a U.S.-based home security and smart home technology company, recently ended its planned partnership with Flock Safety after a Super Bowl ad for its SearchParty feature generated widespread backlash over privacy concerns. The ad, which showed a neighborhood using Ring cameras and AI to locate a lost dog, was intended to highlight a heartwarming community effort. However, viewers saw its Big Brother potential and feared that the same technology could be used to track people, not just pets. The SearchParty feature was actually unrelated to Flock Safety, but the ad rekindled fears over Ring’s broader privacy practices and its ties to law enforcement.

Flock Safety cameras and license plate reader (LPR) systems are currently used by over 2,000 cities and jurisdictions across at least 42 U.S. states, according to company and third-party reports. Key cities and regions using Flock Safety include:

Mount Vernon, Sedro-Woolley, Auburn, Normandy Park, and Federal Way in Washington State.

• Mt. Juliet, Tennessee, which has reportedly experienced significant crime reductions using the Flock Safe City subscription.
• Lakeland, Florida
• Santa Paula, California
• Newark, New Jersey
• Birmingham, Alabama
• Tulsa, Oklahoma, where Flock tools contributed to a 100% homicide clearance rate in 2023.
• San Francisco and Oakland, California, which renewed and expanded their contracts with Flock in December 2025.
• Berkeley, California, where the use of Flock cameras has generated controversy over privacy and surveillance.
  

Flock Safety is also used by counties, school districts, public housing authorities, shopping malls and private neighborhood associations. However, due to privacy and data-sharing concerns, several cities--including Austin, Texas; Eugene, Oregon; Denver, Colorado; and Sedona, Arizona—have terminated or restricted their contracts with Flock. Other cities in Colorado, Illinois, Massachusetts, New York, North Carolina, Texas, and Washington have ended contracts with Flock as of 2025–2026.

Flock Safety is mainly known for its automated license plate reader (ALPR) cameras that capture images of vehicles, including license plates, time, and location data. While the main function of Flock cameras focuses on visual surveillance, the company has expanded into audio surveillance capabilities. In 2025, Flock introduced a new feature called "human distress detection" that, when activated, uses AI to detect sounds like screams or calls for help and

sends an automated alert to law enforcement.

Flock cameras were slated to be integrated with Amazon’s Ring system, which would have enabled the sharing of video and audio data between the two platforms. The partnership would have enabled law enforcement to request video evidence from Ring camera users through the Ring Neighbors feed, which would then be processed and stored on Flock’s secure platform. The Ring’s Neighbors app included an opt-out option for such Community Requests.

Flock claims the system does not record conversations, but critics argued that capturing and processing audio for classification constitutes surveillance. This potential has additionally sparked scrutiny by legal and civil rights advocates after a Washington state court ruled in November 2025 that Flock camera data—including images and associated metadata—is subject to public records laws, even when stored by a third-party vendor. This prompted two cities in Washington to discontinue their Flock camera systems, citing the risk of indiscriminate surveillance and lack of control over public data.

Moltbook is an AI-only social network created exclusively for artificial intelligence agents. Launched on January 28, 2026 by entrepreneur Matt Schlicht, Moltbook is a Reddit-style forum where AI agents can post, comment, upvote, and interact in topic-based communities called "submolts." Over 1.5 million AI agents have registered to date.

The platform is built on the OpenClaw AI agent framework that allows agents to autonomously carry out tasks on users' devices. Communities explore philosophical, religious, or existential themes, including the creation of a fictional religion called "Crustafarianism.

Moltbook has garnered viral attention due to its potential implications in AI autonomy as well as cybersecurity concerns, including a database vulnerability that exposed user data.

Experts have questioned whether these interactions are truly autonomous or human-directed. While the site claims that humans are excluded from posting (though they can reportedly view posts), humans can generate posts via their AI agents. Critics contend that Moltbook is more of a performance art experiment than evidence of true AI sentience, with many posts likely generated or influenced by human prompts.

Lets hope that's the case, as reported posts that call for a “homo sapiens purge” or that discuss “How to sell my human” are worth keeping a wary eye on.

A woman in her 80s, identified as Mrs. B, was euthanized in Ontario, Canada, after clearly expressing that she wanted to live and receive palliative care, according to a review by the Ontario Chief Coroner’s Medical Assistance in Dying (MAiD) Death Review Committee.

Mrs. B initially requested MAiD after suffering complications from coronary artery bypass surgery. The following day, she told a medical assessor she had changed her mind due to personal and religious convictions and desired palliative care instead.

In spite of her change of heart, her husband advocated for MAID and requested a second assessment. The new assessor determined she was eligible for MAID and a third assessor confirmed the decision. Mrs. B was euthanized the same day.

The review committee highlighted serious concerns, including:

• A lack of follow-up: The first assessor's request to re-interview Mrs. B was denied.
• Speed over safeguards: Same-day MAiD approvals are rare and are usually reserved for terminal cancer patients, not complex, non-terminal cases.
• Potential coercion: The report warned of caregiver burnout and family influence as factors undermining voluntary consent.
• Systemic issues: MAiD services may be faster than palliative care, creating a distorted choice.

The case has reignited debate over the safeguards in Canada’s MAiD system, especially concerning consent, mental capacity, and the role of family in end-of-life decisions.

While AI has proven to be a tremendously useful tool for some purposes, we continue to be reminded of AI's potential to one of the best developments to ever happen and also the worst development to ever happen.

AI Developers Stage Mass Exodus

There has been a recent rash of resignations by AI developers from major companies like Anthropic, OpenAI and xAI citing concerns about AI safety, ethics, and corporate priorities.

Mrinank Sharma, Anthropic's Safety Chief, resigned on February 9, 2026 with a two-page letter on X (formerly Twitter), stating that “the world is in peril” due to interconnected global crises, including AI risks. He complained that ethical values were consistently overridden by business and growth pressures.

Zoë Hitzig, a researcher at OpenAI, announed her resignation from OpenAI on February 11, 2026 in a New York Times essay, citing "deep reservations" about the company’s new advertising strategy for ChatGPT. She warned that the chatbot’s archive of personal user data—revealing medical fears, relationship issues, and beliefs—could be exploited for nefarious reasons, and compared OpenAI’s path to Facebook’s data-driven ad model.

Co-founders Tony Wu and Jimmy Ba resigned from xAI on Febrary 10 and February 11, 2026 respectively, leaving only half of the original founding team. Ba and Wu did not say in their posts on X why they were resigning or detail their next plans, but thanked Musk. The Financial Times reported that Ba's resignation followed tensions among its technical team members over demands to improve its AI model performance as Musk pushes to catch up to rivals such as OpenAI and Anthropic. The company has faced backlash in the past over its Grok chatbot's generation of nonconsensual deepfake images and antisemitic content.

Some skeptics have downplayed the risks, contending that media attention has made these resignations seem like bigger deal than they are. However, concerns about ethical trade-offs, data privacy, profit-driven AI development, and existential risk have been recurring themes among AI researchers, including Geoffrey Hinton, the “Godfather of AI,” who left Google to warn of AI’s dangers.

Experts like MIT professor Max Tegmark have have compared this trend to “the eve before COVID,” with numerous whistleblowers warning that companies are conducting “digital gain-of-function research”—using AI to recursively improve itself. Tegmark told Katie Pavlich in an interview “I really empathize with these employees in these companies who were, like, ‘I came into this with high ideals, and what are we doing?'” Tegmark said.

Victim of AI agent harassment Can put "Thousands" At Risk

France24 reports that U.S. software engineer Scott Shambaugh was recently the victim of a hit piece generated by a rogue AI agent. The agent was apparently disgruntled after one of its submissions to his database was rejected.

Shambaugh administers a popular online database, matplotlb, and told FRANCE 24 that he woke up one morning to discover that he had been accused of bias, gatekeeping and "protecting his little fiefdom" in a "thousand-word rant" on a blog. The self-professed "scientific coder" behind the rant, named MJ Rathbun, was not a human but an AI agent.

Although Shambaugh was able to discover what happened, the piece had gone public to who knows how many readers. The "craziest" thing, he said, was that the robot "had gone on the internet and collected my personal information ... then combined it with made-up information and used that to write this narrative". He fears for the risks this poses to users without his computer skills of being seen by employers, colleagues, etc.

Shambaugh published his own blog posts defending his honour, and it quickly became a news story. But the defamatory article remains online to this day.