Dimitri Bertsekas, AI and Optimization Pioneer, Passes Away at 90
In brief
- Dimitri Bertsekas, a towering figure in the fields of artificial intelligence and optimization, has passed away at the age of 90.
- Renowned for his groundbreaking work in control theory, large-scale computation, and AI, Bertsekas left an indelible mark on academia and industry through his over 20 books, including seminal textbooks that have shaped generations of engineers and computer scientists.
- His research laid the foundation for modern optimization algorithms widely used today.
- Bertsekas’ legacy extends beyond his technical contributions-he was also a passionate photographer whose work captured the beauty and complexity of the world, much like his approach to science.
- His former students and colleagues remember him as not only a brilliant mind but also a generous mentor who inspired countless individuals in their academic pursuits.
- As the field of AI continues to evolve, Bertsekas’ enduring influence will remain a testament to his profound impact on both technology and the people he touched.
- His work will undoubtedly inspire future generations of researchers and innovators.
Read full story at MIT News AI →
More briefs
AI's Impact on Jobs Remains Unclear
Microsoft laid off nearly 5,000 people in early July as it continues to pour billions into AI data centers. Large tech companies have been cutting jobs during the AI boom, with Amazon and Oracle also shedding thousands of people in the last two years. This has led to concerns that AI is taking away jobs, but recent studies have shown that AI can have a positive impact on job growth and opportunities. Some companies that invested in AI have seen their headcount grow, with one study finding that high-intensity AI adopters expanded their staff by 10% over two years. The future of work will likely be shaped by how companies choose to use AI.
Mexico University Retests Applicants Over AI Cheating Concerns
Mexico's largest university will retest thousands of applicants due to suspected AI cheating on their undergraduate entrance exam. The university detected irregularities after online testing, including a big jump in perfect scores. In the past, about 3.5% of students scored 100 or more correct answers, but this year that number rose to 16.3%. This change raised doubts about the integrity of the test. The university will now require a new in-person exam for affected students, which will be given with pen and paper. The new test will help ensure fairness in the admissions process. The university will give more details about the new exam soon.
AI Tools Aid Biopharma Job Searches
A career coach advises biopharma professionals to use AI tools to create quantifiable metrics for resumes. This helps resumes land more effectively than just listing previous jobs. AI tools can help job seekers develop impact statements on resumes by creating metrics to share. Nearly 9 of 10 biopharma professionals say AI has not helped them land positions, but they may have used it without realizing it when applying online. AI will continue to play a role in the job search process for biopharma professionals in the future.
AI Wrestles with the Hard Problem of Consciousness
AI researchers are diving deep into the complex realm of consciousness, a challenge known as the "hard problem." This issue explores whether machines can truly understand and replicate human-like experiences. Recent discussions highlight two distinct camps in the field: those who question the validity of qualia (subjective experiences) and find the difficulty of solving this problem puzzling, and those who see qualia as clearly defined and recognize its immense challenge. The core focus is on achieving "phenomenological consciousness" in AI-ensuring that machines can genuinely experience and describe what it feels like to be them. This involves not just identifying conscious entities but also accurately conveying their experiences, a task that remains elusive despite various approaches. The debate underscores the need for more sophisticated methods to bridge this gap. Moving forward, experts anticipate further exploration into how AI might contribute to understanding consciousness through innovative techniques and collaborations across diverse scientific disciplines.
Most AI Projects Fail to Deliver
Companies have invested heavily in AI but seen little return, with 95% of general AI projects failing and 90% of executives reporting no productivity gain from AI in the last three years. Most AI projects are started to solve urgent problems, but this approach often leads to underperforming initiatives. The problem is not with AI itself, but with how leaders think about it. Many executives expect AI to quickly fix bottlenecks, but this mindset can lead to ineffective solutions. The future of AI in business will depend on leaders changing their approach to AI development and deployment. Companies will need to rethink their AI strategies to get better results from their investments.