Yazılar

New Study Finds Major AI Assistants Frequently Misrepresent News Content

/in /tarafından

A new international study from the European Broadcasting Union (EBU) and the BBC has found that leading AI assistants—including ChatGPT, Copilot, Gemini, and Perplexity—misrepresented or mishandled news content in nearly half their responses. The research, published Wednesday, examined 3,000 AI-generated answers to news-related questions in 14 languages, assessing factual accuracy, sourcing, and the ability to distinguish fact from opinion.

The findings were troubling: 45% of AI responses contained at least one significant factual or interpretive issue, while 81% showed some form of problem, ranging from poor attribution to incorrect information. Roughly one-third of all replies featured serious sourcing errors, such as missing or misleading references. Notably, 72% of Google’s Gemini outputs contained significant sourcing flaws—far higher than the under 25% rate for other assistants.

Accuracy issues appeared in 20% of total responses, including outdated or false claims. Examples cited include Gemini incorrectly describing legal changes on disposable vapes, and ChatGPT erroneously identifying Pope Francis as still alive months after his reported death.

The study, involving 22 public-service media organizations across 18 countries, warned that the growing use of AI assistants for news—especially among younger audiences—could threaten public trust. According to the Reuters Institute’s 2025 Digital News Report, 15% of people under 25 now rely on AI assistants for news updates.

“When people don’t know what to trust, they end up trusting nothing at all, and that can deter democratic participation,” said Jean Philip De Tender, EBU’s media director. The report calls for greater accountability and transparency from AI developers to ensure reliable and responsibly sourced information.

Google Unveils Bold AI Upgrades, Premium Plans, and Smart Glasses at I/O 2024

/in /tarafından

At its annual Google I/O conference in Mountain View, California, Alphabet unveiled sweeping updates to its artificial intelligence strategy, including new AI tools, a $249.99/month Ultra subscription plan, and a renewed push into smart glasses.

The announcements come amid intensifying competition from rivals like OpenAI, Anthropic, and Meta, and signal Google’s intent to maintain dominance in search, while rapidly expanding AI services for both consumers and professionals.

AI Mode in Google Search

Google introduced “AI Mode” for U.S. users, transforming traditional search by replacing web links with AI-generated responses for complex queries. Rolled out as an experiment in March, it is now available more widely and aims to deliver deeper, contextual answers.

Gemini App and AI Agents

CEO Sundar Pichai announced that the Gemini AI assistant has reached 400 million monthly active users. The company showcased Gemini’s ability to:

  • Perform real-world tasks like adding events to calendars via smartphone camera scans,

  • Fetch email info and contextual data in conversation,

  • Act as a “universal AI agent” requiring minimal prompting.

AI Ultra Plan – $249.99/Month

Google launched its AI Ultra Plan, targeting power users with:

  • Early access to experimental tools like:

    • Project Mariner – browser automation via keystrokes/mouse clicks,

    • Deep Think – enhanced reasoning with the top-tier Gemini model,

  • 30 TB of cloud storage,

  • An ad-free YouTube subscription.

This premium plan rivals OpenAI’s and Anthropic’s ~$200/month enterprise offerings, and reflects the escalating costs of AI development. It joins Google’s growing portfolio of subscription services, which now count 150 million+ paid users.

Return to Smart Glasses & XR Headset

Google returned to the smart glasses race with new frames running Android XR, demonstrated with live real-time language translation and context-aware Gemini responses during a walk around the I/O venue.

In addition, the company announced:

  • A new XR headset co-developed with Samsung, launching later in 2024,

  • Partnerships with Warby Parker and Gentle Monster to build stylish smart glasses with AI integration.

Search Pressure & Market Outlook

Despite these advances, Alphabet is under pressure. It lost $150 billion in market value earlier this month after testimony revealed that AI had reduced searches in Apple’s Safari browser — a key source of Google’s traffic.

Analysts now warn that Google’s search market share could drop from 90% to under 50% in the next five years due to consumer shifts toward AI chatbots over traditional search.

However, Google sees opportunity in this transition. Executive Robby Stein suggested that more complex AI interactions could lead to new forms of targeted advertising, Google’s main revenue source.

New AI Model – Veo 3

Google also introduced Veo 3, a powerful AI model capable of generating high-quality video and audio, allowing creators to produce realistic, cinematic content through natural prompts.

Investment in AI

Alphabet is going all-in: it’s forecasting $75 billion in capital expenditures for 2025, up from $52.5 billion in 2024, with AI development as the central focus.

Former Meta Executives Raise $15 Million for AI Assistant Startup Yutori

/in /tarafından

Two former Meta artificial intelligence executives, Devi Parikh and Dhruv Batra, have secured $15 million in funding for their AI assistant startup, Yutori. The funding round was led by Radical Ventures’ Rob Toews, with contributions from Felicis, AI pioneer Fei-Fei Li, and Google DeepMind’s Jeff Dean.

San Francisco-based Yutori is developing autonomous AI agents capable of executing tasks independently, a growing trend in AI innovation. Unlike current chatbots that primarily facilitate conversations, Yutori aims to create AI assistants that actively perform tasks, such as managing travel logistics and automating online transactions.

The company is leveraging post-training techniques to enhance AI models’ ability to navigate the web and execute complex reasoning tasks. Yutori’s team includes experts who played key roles in Meta’s AI advancements, including researchers behind Llama 3 and Llama 4, as well as embodied AI projects designed for robotics.