Positive Returns - October, 2025

Oct 2, 2025

Legal Updates: United States - October, 2025

Oct 2, 2025

Policy Updates: October, 2025

Oct 2, 2025

Home-Grown Tech Stacks

Governments are reassessing their reliance on foreign tech, with European countries moving away from U.S. tech giants like Microsoft and scrutinizing foreign-controlled AI apps over data privacy concerns. In response, some companies are promoting digital sovereignty through local platforms or negotiating “sovereignty as a service” deals.

Oct 2, 2025

Who’s Liable for AI Harms?

Lawmakers are struggling to define liability for AI-related harms. The EU dropped its efforts to advance the AI Liability Directive. Federal efforts in the US increasingly shield tech companies from accountability, while states are pushing forward with stronger regulations.

Oct 2, 2025

Paying Upfront for Space Travel

Oct 2, 2025

Q-Day is Coming. Are You Ready?

Microsoft signaled a major step toward quantum-powered cloud computing with early workloads expected by 2027. This raises urgent cybersecurity concerns, and underscores the importance of strengthening data protection now and planning for post-quantum cryptography.

Oct 2, 2025

Are We Ready for “the Merge?”

Brain-computer interface companies like Neuralink and Merge Labs are advancing invasive technologies that connect the brain directly to external devices, raising major ethical, legal, and privacy concerns.

Oct 2, 2025

Preempting Freedom

New digital tools enable law enforcement to conduct widespread surveillance and predictive profiling, raising serious civil liberties concerns and fueling fears of growing authoritarianism through “preemptive security” tactics.

Oct 2, 2025

Dual-Use, Unlimited Risk?

VC investment in dual-use tech is booming as startups aim to corner both government and private markets. As dual-use tech proliferates, concerns are growing over the ethical implications of AI-powered military systems like the “kill cloud,” with companies like Microsoft facing backlash for enabling surveillance and human rights abuses by militaries.

Oct 2, 2025

Prompt Injection Needs Prompt Attention

LLMs and AI coding agents are increasingly vulnerable to “prompt injection” attacks that can trick them into revealing sensitive data or executing malicious code. This raises significant security risks as these tools become more common and gain more authority in development environments.

Oct 2, 2025

Meta Leads the Race to the Bottom

Meta’s recent policy rollbacks and lack of transparency have led to increased harmful content and user distrust, and the company’s automation of content moderation raises concerns about safety and integrity. Several other social media platforms are nevertheless following Meta’s lead.

Oct 2, 2025

Going Nuclear

To meet AI’s growing energy needs while reducing carbon emissions, companies are investing in advanced nuclear technologies, including small modular reactors and fusion power. But natural gas remains the dominant energy source for data centers today, raising concerns about overreliance on fossil fuels and the need for greater transparency in energy sourcing.

Oct 2, 2025

Assessing AI’s Energy Appetite

The true environmental cost of AI is vastly underestimated, and the companies developing it remain opaque about energy use and emissions. Despite some efforts at transparency, the current pace and scale of AI deployment is unsustainable without clearer accountability and better efficiency metrics.

Oct 2, 2025

To Replace or Redeploy?

Companies are taking varied approaches to AI adoption, with some opting for job cuts, while others are redeploying workers into AI-focused roles. However, experts caution that current AI tools still lack key capabilities, and premature workforce replacement could backfire if companies overestimate what AI can actually do.

Oct 1, 2025

What’s Really Taking White Collar Jobs?

The recent surge in unemployment among college graduates is driven by a mix of factors, not just AI. While AI is augmenting tasks rather than replacing most jobs, companies appear to be scaling back entry-level hiring and training in anticipation of future automation, making it harder for early-career workers to enter the job market.

Oct 1, 2025

Authoritarian AI

AI is accelerating the spread and sophistication of global disinformation, with chatbots, video generators, and social media platforms being used to manipulate public opinion, disrupt elections, and amplify propaganda—often at the hands of authoritarian regimes.

Oct 1, 2025

The AI Kids Are Not Alright

AI systems accessible to children must prioritize youth safety, transparency, and privacy from the outset. Yet companies like Meta and some toy makers are already ignoring these guidelines, raising serious concerns about exploitation, inappropriate content, and unknown developmental risks for children.

Oct 1, 2025

AI is Listening, and it Can’t Keep a Secret

AI companies are increasingly harvesting vast amounts of personal data to fuel model development and personalized services, which raises serious privacy, security, and legal concerns. As generative AI faces technical limits, experts warn that firms may shift to surveillance-based business models, exploiting sensitive user data without meaningful consent or transparency.

Oct 1, 2025

A World Beyond LLMs?

A new Apple study confirms that LLMs fail at true reasoning, relying on pattern matching rather than logic. Experts warn that current evaluation methods overstate LLM capabilities, and some researchers argue that advancing AI will require moving beyond LLMs toward other approaches that better mimic human understanding.

Oct 1, 2025

AI Bubble Watch

Despite massive VC investment, many AI startups are struggling to deliver returns, with valuations skyrocketing even as most projects fail to generate meaningful revenue and the costs of model training and inference rise. Experts are cautioning that the market may be more overvalued than in the dot-com era.

Oct 1, 2025

All Eyes on Agents

AI agents are generating major interest, with both Big Tech and VC rapidly investing in the space. But many so-called "agents" are just rebranded chatbots or automation tools. Despite their promise, real-world performance remains inconsistent, and concerns over reliability, misuse, and integration challenges are leading some analysts to warn of high failure rates.

Oct 1, 2025