Wired

No, Sam Altman, AI Won’t Solve All of Humanity’s Problems

Metaculus

When will we have transformative AI?

Elon Musk

Yoshua Bengio

Demis Hassabis

Melanie Mitchell

Stuart Russell

Sam Altman

Dario Amodei

Geoffrey Hinton

Yann LeCun

Gary Marcus

Max Tegmark

Shane Legg

Society is more likely to get "cat-level" or "dog-level" AI years before human-level AI.

"[W]e will eventually reach AGI... and quite possibly before the end of this century."

"Superhuman AIs" will be achieved "between a few years and a couple of decades."

"Digital intelligence" could overtake us within "5 to 20 years."

We may achieve "something like AGI or AGI-like in the next decade."

"[W]e could get to real AGI in the next decade."

"Human-level" AI could occur in "two or three years."

"I'd be surprised if we don't have AGI by [2029]."

Will AI Surpass Human Intelligence, and When?

An advanced version of Google Gemini 2.5 Deep Think solved 10 out of 12 problems at the 2025 International Collegiate Programming Contest (ICPC) World Finals in Azerbaijan, winning a gold medal and ranking second overall among 139 competing university teams.

Financial Times

The Guardian

Google

An advanced version of Google Gemini 2.5 Deep Think solved 10 out of 12 problems at the 2025 International Collegiate Programming Contest (ICPC) World Finals in Azerbaijan, winning a gold medal and ranking second overall among 139 competing university teams.

The ICPC is considered the world’s most prestigious algorithmic programming competition at the college level, with teams from over 3,400 universities across 103 countries competing in its latest edition.

Worldfinals

The ICPC is considered the world’s most prestigious algorithmic programming competition at the college level, with teams from over 3,400 universities across 103 countries competing in its latest edition.

Gemini 2.5 Deep Think competed remotely, following ICPC rules, and solved eight problems within 45 minutes, and completed two more within three hours, for a combined total time of 677 minutes.

Gemini 2.5 Deep Think competed remotely, following ICPC rules, and solved eight problems within 45 minutes, and completed two more within three hours, for a combined total time of 677 minutes.

Notably, the AI model was able to solve Problem C within 30 minutes (a complex liquid distribution optimization challenge that no human team could complete) using dynamic programming algorithms and nested ternary searches to find the optimal solution.

Notably, the AI model was able to solve Problem C within 30 minutes (a complex liquid distribution optimization challenge that no human team could complete) using dynamic programming algorithms and nested ternary searches to find the optimal solution.

In a press release, Google described the achievement as demonstrating “a profound leap in abstract problem solving,” with Google DeepMind Vice President Quoc Le comparing it to previous AI breakthroughs such as Deep Blue in 1997 and AlphaGo in 2016.

In a press release, Google described the achievement as demonstrating “a profound leap in abstract problem solving,” with Google DeepMind Vice President Quoc Le comparing it to previous AI breakthroughs such as Deep Blue in 1997 and AlphaGo in 2016.

Google was only bested by OpenAI’s GPT-5, which topped the ICPC rankings with a perfect score after solving all 12 problems, including 11 correctly on the first attempt and the most challenging after the ninth submission.

Mostafa Rohaninejad

Openai

Google was only bested by OpenAI’s GPT-5, which topped the ICPC rankings with a perfect score after solving all 12 problems, including 11 correctly on the first attempt and the most challenging after the ninth submission.

Wired #1

Metaculus #1

OpenAI's GPT-5 model achieved a perfect score by solving all 12 problems at the same competition, with 11 problems solved correctly on the first attempt and only requiring 9 submissions for the most difficult problem.</span

VentureBeat

VentureBeat #

Financial Times #

OpenAI's GPT-5 model achieved a perfect score by solving all 12 problems at the same competition, with 11 problems solved correctly on the first attempt and only requiring 9 submissions for the most difficult problem.

The International Collegiate Programming Contest is recognized as the world's most prestigious algorithmic programming competition at the college level, with teams from nearly 3,000 universities across 103 countries competing annually.</span

Google #

The International Collegiate Programming Contest is recognized as the world's most prestigious algorithmic programming competition at the college level, with teams from nearly 3,000 universities across 103 countries competing annually.

Gemini successfully solved Problem C within 30 minutes, a complex liquid distribution optimization challenge that no human team could complete, using dynamic programming algorithms and nested ternary searches to find the optimal solution.</span

The Guardian #5507

Gemini successfully solved Problem C within 30 minutes, a complex liquid distribution optimization challenge that no human team could complete, using dynamic programming algorithms and nested ternary searches to find the optimal solution.

The competition requires teams to solve complex algorithmic problems within a five-hour time limit, with rankings based on accuracy, speed, and number of problems solved, where only perfectly correct answers receive points.</span

The competition requires teams to solve complex algorithmic problems within a five-hour time limit, with rankings based on accuracy, speed, and number of problems solved, where only perfectly correct answers receive points.

Google DeepMind described this achievement as a historic moment toward artificial general intelligence, comparing it to previous AI breakthroughs like Deep Blue defeating chess champion Garry Kasparov in 1997 and AlphaGo beating the world Go champion in 2016.</span

Google DeepMind described this achievement as a historic moment toward artificial general intelligence, comparing it to previous AI breakthroughs like Deep Blue defeating chess champion Garry Kasparov in 1997 and AlphaGo beating the world Go champion in 2016.

The enormous computing power required for these achievements raises questions about practical applicability, as competitive coding success doesn't necessarily translate to real-world software development where problems take months to solve rather than hours. The pressure on AI companies to claim breakthroughs is enormous, and these controlled competition environments may not reflect the collaborative, long-term thinking required for actual engineering challenges.

The enormous computing power required for these achievements raises questions about practical applicability, as competitive coding success doesn't necessarily translate to real-world software development where problems take months to solve rather than hours. The pressure on AI companies to claim breakthroughs is enormous, and these controlled competition environments may not reflect the collaborative, long-term thinking required for actual engineering challenges.

DeepMind and OpenAI achieve gold at ‘coding Olympics’ in AI milestone

Artificial Intelligence

AI Models Win Gold at World Programming Competition

Headlines

These AI victories represent a transformative leap toward artificial general intelligence that will revolutionize scientific and engineering disciplines. The genuine creativity and abstract reasoning demonstrated by these models prove that AI can serve as a true problem-solving partner for programmers and researchers.

These AI victories represent a transformative leap toward artificial general intelligence that will revolutionize scientific and engineering disciplines. The genuine creativity and abstract reasoning demonstrated by these models prove that AI can serve as a true problem-solving partner for programmers and researchers.

Gemini achieves gold-level performance at the International Collegiate Programming Contest World Finals

Although AI is a powerful tool, which demonstrated its prowess at ICPC 2025, those who wield it must remember that it cannot solve every problem the world currently faces. Many of these obstacles, significantly, are man-made and require a human-led and engineered response to solve them that AI can never replicate.

Although AI is a powerful tool, which demonstrated its prowess at ICPC 2025, those who wield it must remember that it cannot solve every problem the world currently faces. Many of these obstacles, significantly, are man-made and require a human-led and engineered response to solve them that AI can never replicate.

We will have transformative AI by Dec. 2040, according to the Metaculus prediction community.

We will have transformative AI by Dec. 2040, according to the Metaculus prediction community.

Google’s Deep Think Wins Gold at World Programming Competition

AIGoogle’s ModelsDeep WinThink Wins Gold at World Programming Competition

The Spin

Metaculus Prediction

The Controversies

LeCun

Marcus

Bengio

Hinton

Hassabis

Altman

Amodei

Musk

Go Deeper

Articles on this story

Sign Up for Our Free Newsletters
Sign Up for Our Free Newsletters